Full Report

The European Used-Car Industry — Understanding the Playing Field

Figures converted from EUR at historical FX rates — see data/company.json.fx_rates for the rate table. Ratios, margins, and multiples are unitless and unchanged. Pan-European market-size figures are kept in their source currency (EUR), as sourced, and are not converted.

AUTO1 Group is a Berlin-based digital platform that buys used cars from consumers, wholesales most of them to professional dealers, and retails a growing minority directly to other consumers — all across roughly 30 European countries. To judge AUTO1 you first have to understand the arena it competes in: a vast, old, intensely fragmented, and barely-digitised market where the money is made one car at a time, on margins so thin that scale and cost-per-car decide who wins. This tab builds that mental model from the primary record, then frames the competitive, cyclical, and regulatory forces an investor needs before reading the rest of the report.

The single most important idea: this is not a high-margin software business. It is a physical, principal-trading business — AUTO1 takes title to the cars on its own balance sheet — wrapped in technology. Gross margin at the group level is about 12% [10]. Everything in the industry flows from that fact.

1. The arena in one minute

Sources: AUTO1 Capital Markets Event 2026 [1] [2]; Q4/FY2025 earnings call [3].

Europe's used-car market is worth roughly €700 billion a year, with an additional €100 billion financing pool layered on top [1]. About 27.5 million used cars change hands annually across the continent [3]. To frame the scale: used-car sales account for almost two-thirds of all automotive transactions in Europe — the used market is far larger by unit count than the new-car market [5].

Two structural features define the playing field:

• Extreme fragmentation. There are more than 250,000 used-car dealers in Europe, and the top 20 own less than 6% of the market [2]. No incumbent has scale; the market is a long tail of small local lots. That is the gap a pan-European platform is built to consolidate.
• Barely digitised. Management describes used cars as "the largest vertical with a low online share," still sold "almost exclusively offline" [4]. The structural thesis for the whole sector is a slow migration of these transactions from offline to online — and 77% of consumers say they dislike the current car-buying experience [2].

Against that backdrop, AUTO1's 3.1% share in 2025 — with a stated long-term target of 10% — tells you how early this consolidation still is [3].

2. How the money is actually made: three doors into one platform

A newcomer's first job is to learn the three customer-facing motions and the jargon attached to each. AUTO1 runs one vertically integrated platform with three brands [7]:

• C2B sourcing (consumer-to-business) — the wirkaufendeinauto.de brand. Consumers sell their old car to AUTO1 at a haggle-free price, often dropping it at a nearby branch. This is the supply funnel that feeds everything else.
• B2B wholesale — the "Merchant" segment — AUTO1.com, Europe's #1 online wholesale platform, where 54,000 professional dealers buy cars in digital auctions across 30+ countries [8]. Started in 2013, this is the larger, profitable engine [9].
• B2C retail — the "Retail" segment — Autohero, a direct-to-consumer online used-car store (refurbished cars, fixed prices, home delivery, 21-day returns). Started in 2020, it is the high-growth, still-loss-making disruptor [9].

Three terms you must internalise to read this sector:

Principal vs. marketplace. AUTO1 is a principal trader — it buys the car onto its own balance sheet and resells it, so revenue is the full car price and the company carries inventory and price risk. A marketplace or classifieds business (e.g. Auto Trader) only takes a listing fee and never owns the car. This single distinction explains nearly all the margin differences across the peer group below.

GPU — gross profit per unit. Because every car is a transaction, the industry is run on dollars of gross profit per car sold, not on percentage margins. GPU is the master KPI. Add up GPU across cars, subtract cost-per-car, and you have the business.

ASP — average selling price. The average price of the cars traded (roughly $9,000–10,500 in Merchant, higher in Retail). ASP drifts up over time as the fleet ages into more expensive model years [3].

The reason a platform can beat 250,000 local dealers is vertical integration at continental scale: a proprietary dataset of 6+ million transactions feeding AI pricing, 12 production (refurbishment) centres with capacity for 248,000 cars, 170+ logistics centres, and 750+ drop-off branches — the largest cross-border car-logistics network in Europe, built over more than a decade [7] [18]. A car with a weight of one to two tonnes "cannot be conquered by an AI prompt" — the physical network is the moat [10].

The two engines have opposite economics

Source: AUTO1 Capital Markets Event 2026 — Merchant per-unit economics [11] and Retail per-unit economics [12].

This table is the whole industry in miniature. Merchant is a high-volume, low-GPU machine that already makes money per car ($380 of adjusted EBITDA per unit in 2025). Retail carries a far higher GPU ($3,100) but also a far heavier per-car cost base — refurbishment, brand marketing, last-mile delivery — and still loses $482 per car, a loss management is deliberately narrowing each year as scale spreads fixed costs. The long-term targets imply both engines turning into $470–940 of EBITDA per car, which is the entire bull case in one row [14].

3. Why it is a thin-margin, scale-driven business

In 2025 AUTO1 sold 842,000 cars, generated $1,164 million of gross profit (up 37%), and turned that into $233 million of adjusted EBITDA — an adjusted EBITDA margin of just 2.4% [10]. Group gross profit per unit was $1,377 [10]. On a car selling for several thousand dollars, that is a wafer-thin slice — and it is why the industry is so unforgiving: there is no room for sourcing mistakes, mispricing, or logistics waste. The companies that survive are the ones that grind cost-per-car down faster than competitors and attach high-margin extras (financing, warranties) to each sale.

The structural promise is operating leverage: most of the cost base is fixed (refurbishment centres, logistics, technology, brand), so as volume grows, cost-per-car falls and a rising share of each incremental GPU drops to EBITDA. AUTO1's own evidence: gross profit grew 37% while adjusted EBITDA grew 81% in 2025 [10], and management targets a long-term adjusted EBITDA margin of 5% to 9% versus 2.4% today [3]. The whole investment debate is whether that leverage is real or whether thin margins are a permanent feature of trading physical cars.

The clearest way to see the engine working is the multi-year GPU climb — the rising gross profit captured on every car, which comes from better AI pricing, more attached financing, and value-added services:

Source: AUTO1 Capital Markets Event 2026 — Merchant GPU track record [11] and Retail GPU track record [12]; converted at each fiscal year's period-end EUR/USD rate.

Merchant GPU rose from $848 to $1,147 over five years (long-term target $1,410+); Retail GPU climbed seven-fold off a tiny base to $3,100 (long-term target $3,525+) [11] [12]. A meaningful and growing slice of GPU now comes from embedded financing — a captive lending layer attached to each car that carries far higher margins than the metal itself, and which management views as the long-term profit unlock [10].

4. Market size, fragmentation, and the offline-to-online runway

How big is the opportunity, really? The market can be sized as a funnel, from the total European car fleet down to the slice a platform can realistically capture:

Source: AUTO1 Capital Markets Event 2026 — Europe-wide sourcing opportunity [16].

Continental Europe has a car parc of ~190 million vehicles; about 24% of owners intend to sell within 12 months (~45m cars), of which roughly 10–15 million are addressable consumer-to-business selling demand annually [16]. On the demand side, AUTO1 frames ~10 million units a year of external dealer-sourcing demand for its wholesale product — of which it captures only 7% today — and a 15-million-unit addressable retail market for Autohero [15] [17]. Whichever lens you use, AUTO1's current penetration is low single digits — the runway is the story.

The offline-to-online shift has been the sector's structural tailwind since the IPO era: at the 2021 listing, management sized the European used-car market at ~€600 billion in 2019, growing at a 5% CAGR to 2025, and emphasised that it remained "highly fragmented and at a very early stage of online penetration" [6]. Five years on, the same fragmentation persists — confirming that consolidation in this industry is a multi-decade, not multi-year, process [2].

5. Where we are in the cycle

Used-car volumes are mature and mildly cyclical — they oscillate in a band rather than grow secularly. Across Europe, annual transactions have moved between roughly 24 and 28 million over a decade:

Source: AUTO1 Capital Markets Event 2026 — European used-car market transactions [2].

The cycle's defining variable is not unit volume but used-car prices, which feed directly into ASP, GPU, and inventory risk. The most instructive recent episode: the 2021–2022 semiconductor shortage throttled new-car production, pushed buyers into the used market, and sent used-car prices sharply higher — a demand surge that drained dealer inventories to about 17% below pre-pandemic levels [19]. That spike inflated GPUs in 2021, then normalised as supply chains healed — visible in the 2022 volume trough above. Today the market is in a stable phase: volumes grew about 1% in 2025, while AUTO1's own ASP rose ~7% as it trades progressively newer, more expensive model years [3]. The key cyclical lesson for an investor: a principal trader's earnings are levered to price, and a sharp downturn in used-car values would compress GPU and create inventory write-down risk even if unit volumes hold.

For 2026, management guides to 940,000 to 1,000,000 group units, $1.26–1.38 billion of gross profit, and $287–315 million of adjusted EBITDA, all funded from a balance sheet with ~$705 million of cash and no corporate debt — inventory is financed through asset-backed securitisation rather than corporate borrowing [20].

6. The competitive landscape

There is no European pure-peer at AUTO1's scale running the identical dual wholesale-plus-retail model — at IPO the company noted it sold "approximately twice as many used cars as its closest peer in the EU" [5]. The useful way to map the field is by business model, because model — not geography — drives the economics. Three archetypes compete for the used-car profit pool:

• Principal online/omnichannel retailers (take title, sell to consumers): Carvana in the US — "the leading e-commerce platform for buying and selling used cars" [22]; CarMax, "the nation's largest retailer of used vehicles" (780,684 retail units) [23]; and Aramis Group in Europe (France/Spain/Belgium/etc.), the closest listed analogue to Autohero, which sold ~119,000 cars to individuals and runs France at "profitability close to 5% EBITDA" [26].
• Wholesale digital marketplaces (mostly fee-based, the AUTO1.com analogue): OPENLANE, "a leading digital marketplace for wholesale used vehicles" with $28.8 billion of GMV across ~1.5 million transactions [24]; and ACV Auctions, a US dealer-to-dealer wholesale marketplace still running operating losses (–$63 million in 2025) [25].
• Classifieds / advertising marketplaces (never touch the car): Auto Trader in the UK, whose strategy is simply to be "the best place to buy and sell a car" as a listings platform [27].

The spectrum of gross margins makes the model differences unmistakable — and explains why AUTO1's ~12% looks low against software but is entirely normal for a principal car trader:

Sources: latest annual filings — Carvana FY2025 [22]; CarMax FY2026 [23]; Aramis FY2025 [26]; Auto Trader FY2025 [27]; margins as reported.

The four principal traders cluster at 10–21% gross margin — they buy and resell metal. Auto Trader, which only sells listings and never owns a car, earns ~77% gross margin and ~62% operating margin [27]. That gap is the central trade-off of the industry: principal models capture the whole transaction (and its risk) but on thin margins; asset-light models earn fat margins but cede the transaction itself. AUTO1's bet is that owning the physical pipe — sourcing, pricing, refurbishment, logistics, financing — builds a deeper, more defensible position than a listings page, even at one-sixth the gross margin.

Sources: latest reported annual filings — AUTO1 FY2025 [10]; Carvana FY2025 [22]; CarMax FY2026 [23]; Aramis FY2025 [26]; OPENLANE FY2025 [24]; Auto Trader FY2025 [27]. AUTO1 and Aramis converted EUR to USD at the FY2025 period-end rate; US peers reported in USD; Auto Trader shown in GBP (no EUR-based FX rate available). OPENLANE and Auto Trader revenue is fee/advertising revenue, not vehicle value, so gross margin is not comparable to the principal traders.

A note on AI as a competitive force, since investors will ask: management argues large language models and AI-driven search may erode classifieds traffic — the first search touch-point — but cannot replicate a principal trader's physical network, balance sheet, and proprietary transaction-pricing data. As of the latest call, management reports no new entrants attacking the sourcing or retail side [10]. The competitive risk for the classifieds model may be greater than for the principal model.

7. Regulation and structural risks

The European used-car industry is lightly regulated relative to financial or healthcare sectors, but exposed to a mosaic of local rules that shift the goalposts country by country. AUTO1's own risk reporting frames the key channels [21]:

• Vehicle-type / emissions policy. Management explicitly flags that new regulations "can lead to a decline in certain vehicle types, including those in our inventory," which "can negatively affect our margins" [21]. Diesel bans, low-emission zones, and the EV transition can abruptly reprice whole tranches of the used fleet a trader holds — the most material regulatory risk in this business.
• Cross-border friction. Operating across 30+ countries means VAT, registration, and consumer-protection regimes differ in every market; a platform's pan-European logistics has to absorb that paperwork as a cost of doing business.
• Financing → conduct regulation. As AUTO1's captive lending arm grows, it brings the business into anti-money-laundering and data-protection compliance, and consumer-credit conduct rules — a heavier regulatory layer than pure car trading [21].
• Demand-side friction (not regulation, but structural). A persistent brake on the offline-to-online thesis is that some buyers still will not purchase a car they cannot see and test-drive in person; traditional dealers retain a "relatively strong position" in retail [21]. This is why online penetration has climbed slowly despite a decade of investment.

Net assessment: regulatory risk is medium — no single rule threatens the model, but the EV/emissions transition and the financing build-out are live, margin-relevant policy exposures worth monitoring.

8. Watchlist — the signals that would change the industry view

Sources: AUTO1 Q4/FY2025 earnings call [3] [10]; Capital Markets Event 2026 GPU track record [11] [12]; FY2021 industry conditions [19].

The bottom line for a newcomer. The European used-car industry is a €700bn, fragmented, mostly-offline market in the early innings of digitisation [1]. It rewards whoever builds the lowest-cost, largest-scale, vertically integrated machine for moving cars — but it does so on per-car margins thin enough that execution, pricing accuracy, and cost discipline are everything. The structural runway is real and decades long; the day-to-day reality is a capital-intensive, cyclically price-exposed, low-margin trading business. Hold both truths at once, and the rest of this report will read clearly.

---

Figures converted from euros at historical FX rates — see data/company.json.fx_rates for the rate table. Ratios, margins, and multiples are unitless and unchanged.

AUTO1 Group: Two Engines and an Emerging Bank, Priced as One Blurred Average

The one-sentence verdict. AUTO1 is a thin-margin, capital-light-on-the-surface-but-capital-hungry-underneath principal car trader that has reached a genuine profit inflection — and the most useful thing you can do as an investor is stop looking at the consolidated numbers, because the group average hides a profitable, scaling wholesale marketplace (Merchant), a still-loss-making but rapidly-narrowing online retailer (Autohero), and an embedded consumer-and-dealer lending book that is only now starting to matter.

The industry tab already established the arena: a $800bn, fragmented, barely-digitised European used-car market where money is made one car at a time on razor-thin per-unit margins [2]. This tab does the next job: explain how AUTO1 itself makes money, why its returns on capital just turned positive after years of deep losses, where the real value sits, and how an intelligent investor should underwrite it. The short version of the valuation answer: price-to-earnings is the wrong lens; this is a units × gross-profit-per-unit × operating-leverage story best valued on forward EBITDA and a sum-of-the-parts — and on that basis it is demanding on today's results but reasonable on the trajectory, provided you believe management can execute.

Source: Q4/FY2025 earnings call — 842,000 units (+22%), $1,164m gross profit (+37%), $233m adjusted EBITDA (+81%), group GPU $1,377 [1].

1. The economic engine: a spread captured per car, multiplied by scale

Strip away the technology language and AUTO1 is a principal trader that earns a gross-profit spread on every car it touches, then tries to spread a largely fixed cost base across ever-more cars. Management's own framing — "an AI-enabled Amazon for the used car market" — is marketing, but the underlying machine is real: a proprietary dataset of 6M+ transactions accumulated over 14 years feeds AI pricing; a physical network of 170+ logistics centres, 759 drop-off branches and 12 production centres with 248,000 units of refurbishment capacity moves the metal [11].

The profit identity that governs everything is simple:

Adjusted EBITDA ≈ Units × (Gross profit per unit − Operating cost per unit)

Both terms move in AUTO1's favour as it scales. In 2025 group GPU rose 12% to $1,377 while the cost base was spread over 22% more units — so gross profit grew 37% but adjusted EBITDA grew 81% [1]. That gap between gross-profit growth and EBITDA growth is the operating leverage, and it is the entire bull case in one number. The constraint is equally simple: on a car selling for several thousand dollars, $1,377 of group gross profit is a wafer-thin slice, so a small slip in pricing accuracy, logistics cost, or inventory risk wipes out the spread. This is a business with high operating leverage and no margin for error.

The inflection is the story

For most of its public life AUTO1 lost money heavily. The turn from deep operating losses to positive operating income across 2022→2025 — driven entirely by the leverage above, not by a margin trick — is the single most important fact on this page.

Sources: revenue and gross profit from company filings as reported; FY2024–25 adjusted EBITDA of $113m and $233m from Q4/FY2025 earnings call [1]; FY2022–23 adjusted EBITDA derived from segment adjusted-EBITDA disclosures [3] [4]. Operating income as reported.

The same inflection shows up in returns on capital, which crossed from deeply negative to positive only in the last two years — a fragile but real change:

Source: derived from reported financials, FY2022–FY2025 (ROE and ROCE computed from net income, operating income, equity and capital employed); company filings, as reported.

A word of caution on those returns: AUTO1's equity base is thin ($831m) and carries an accumulated deficit of roughly $1.5bn from prior losses, so a 2025 ROE of 11% is flattered by a small denominator and balance-sheet leverage. ROCE of 6.8% is the more honest read of underlying capital productivity — positive, improving, but still below a double-digit cost of capital. This is a business climbing out of a hole, not a proven high-return compounder yet.

2. Why you must un-blend the segments

The consolidated 2.4% EBITDA margin — earned on $9.6bn of 2025 revenue [22] — is an average of two opposite economic profiles. Looking at them separately is the only way to value the company correctly.

Sources: Capital Markets Event 2026 — Merchant segment financials (FY2025 adj. EBITDA $281m, 3.7% margin) [3] and Retail segment financials (FY2025 adj. EBITDA −$49m, −2.4% margin) [4].

Merchant (AUTO1.com — B2B wholesale) is the profit engine and roughly 78% of revenue. It is a digital wholesale marketplace where ~54,000 professional dealers buy cars; it has compounded units at a 51.7% rate since 2013 and now earns $281m of adjusted EBITDA at a 3.7% margin — and crucially, its margin is still expanding (from 3.1% to 3.7% in one year) as GPU rises and per-unit cost falls [18] [3]. This is a profitable, growing, #1-in-Europe asset that, on its own, would screen as a decent marketplace business.

Retail (Autohero — B2C online) is the growth option. It has grown at a 58.5% rate since its 2020 launch and lost money every year — but the loss is narrowing dramatically per car: from −$4,640 of adjusted EBITDA per unit in 2021 to −$482 in 2025, with management saying full-year 2025 retail unit economics were already positive before marketing spend [6]. The entire question for Autohero's value is whether the remaining −$482 closes and turns positive as scale spreads its heavy fixed cost (refurbishment, brand, last-mile delivery).

The per-unit profit-and-loss is where the two models become legible

Sources: Capital Markets Event 2026 — Merchant per-unit economics [5]; Retail per-unit economics [6].

Read this table slowly, because it is the whole investment debate. Merchant captures a small GPU ($1,147) against a small cost base ($767) and already nets $380 per car — and management targets $550–825 long-term. Retail captures a far larger GPU ($3,100) but carries a far heavier cost base ($3,581 of SGA — five times Merchant's per-car marketing alone), so it still loses $482 per car today, with a long-term target of $1,660–2,760 of EBITDA per unit. If Retail merely reaches the low end of its target on its long-term unit goal of 300k cars, that is ~$500m of EBITDA from a segment losing money today. That swing — not the wholesale business — is the source of most of the upside and most of the risk.

3. The third engine almost nobody underwrites: captive finance

The least-appreciated part of AUTO1 is the embedded lending layer management is building into both segments — and it is where the highest-quality incremental margin lives, because interest income carries far better economics than trading metal.

On the consumer side, in-house financing already contributes $469 of Autohero's GPU (up from $154 in 2021), split between an external referral kickback ($222) and captive internal interest ($247), with a stated long-term captive contribution of roughly $1,000 per car [12]. The unit economics management discloses are genuinely attractive for a lender: a net interest margin of ~5% (target 5–7%), an attachment rate of 40% (target 50–60%), and cost of credit of just 1.2% [13].

On the dealer side, Merchant financing scaled fast — attach rate up to 17% (target 50%), with $682m of financed sales (+74%) and 124,000 vehicles financed in 2025 — and it earns a net interest margin of around 7.5% [19] [21]. The risk flag worth keeping: management took a $13.9m Merchant-finance impairment in 2025 from a one-off underwriting lapse in a single newly-launched market — a reminder that scaling a captive lender invites credit risk, not just margin [20].

Source: Capital Markets Event 2026 — Retail Finance GPU, captive internal interest vs external kickback [12].

The strategic logic: financing turns Autohero from a one-off transaction into a multi-year relationship, lifting lifetime value and stickiness [8]. The catch for valuation: a loan book consumes capital, and growing it is a real call on AUTO1's equity even when ABS funds most of it (see §4). A buy-side reader should value this as an emerging captive-finance arm with its own NIM-and-loss-rate economics — not fold it invisibly into a retail multiple.

4. The cash-flow question you must resolve before you own this

Here is the trap. On an IFRS basis AUTO1 burns cash: 2025 operating cash flow was −$544m and free cash flow −$575m. Yet management insists the business "self-funds" its growth and ended the year with ~$705m of cash and no corporate debt [8]. Both statements are true, and reconciling them is the core analytical task.

The bridge is asset-backed securitisation (ABS). AUTO1 funds its two growing asset pools — car inventory and the captive finance book — through non-recourse ABS vehicles at an 80%+ advance rate, which IFRS records as financing inflows while booking the asset build as an operating outflow [10]. Management's preferred measure, "AUTO1 Cash Flow," nets the ABS funding against the asset growth and was positive $76m in 2025:

Source: Capital Markets Event 2026 — AUTO1 Cash Flow walk for FY2025 (Adj. EBITDA $233m to AUTO1 Cash Flow $76m, after net inventory −$46m and net captive finance −$76m post-ABS) [9]. Bridge simplified; intermediate items grouped.

The honest read. The trading business is genuinely self-funding — inventory growth is overwhelmingly covered by ABS, capex is trivial (~25bps of revenue) [8]. The real call on equity is the captive finance book: AUTO1 retains the first-loss equity tranche of its ABS (5–16% today, with a stated target of 1–2%), and growing the loan book outpaced its ABS funding by ~$76m in 2025 [9]. So "self-funding" is fair for the car business but optimistic for the finance build — which is precisely why a stable ~$705m cash cushion and zero corporate debt matter. Verdict: the cash burn is mostly an accounting artefact of growth funded by working-capital-like ABS, not a going-concern problem — but the finance-book equity drag is real and should be watched as the loan book scales. A reader who treats the ~$1.8bn of ABS as corporate debt will badly misjudge both leverage and enterprise value (see §6).

5. The moat: real, mechanical, and unusually physical — but it caps margins, it doesn't lift them

AUTO1's competitive advantage is genuine, and management can articulate the mechanism rather than just assert a brand. The moat rests on three reinforcing pillars, each defensible for a specific reason:

Sources: Q4/FY2025 earnings call — proprietary pricing data, physical network as moat, "cannot be conquered by an AI prompt" [1]; Capital Markets Event 2026 — infrastructure scale (6M+ transactions, 248k capacity, 170+ logistics) [11] and Autohero brand metrics [18].

The most credible moat argument management makes is about data exclusivity: AI pricing needs transaction prices, which are private and only generated by trading; even classifieds that aggregate the market hold only asking prices [1]. Combined with a physical network built over a decade, this is a real barrier — and management reports no new entrants attacking the sourcing or retail side, while arguing that large language models threaten classifieds (the search touchpoint) far more than a principal trader [1]. AUTO1's own FY2022 report is more candid about the competitive reality, calling the sector "highly competitive" with independent dealers, classifieds, and rental fleets as rivals — though it also notes that two well-funded online challengers, Cazoo and Carnext, exited most of AUTO1's markets in 2022 [17]. That failure of better-funded pure-play retailers, while AUTO1 survived and scaled, is the strongest real-world evidence that its integrated model is durable.

The critical nuance: this moat protects volume and survival, but it does not turn AUTO1 into a high-margin business. Owning the physical pipe is what lets it consolidate share — but a principal trader's structural margin is set by the thin spread on metal. The moat makes the volume defensible; it does not change the fact that the economics only work at scale. That is why the moat and the operating-leverage thesis are the same thesis.

6. How to value it — and why the headline P/E is a trap

At $28.0 per share (June 2026) AUTO1 trades at roughly 68× trailing earnings — a number that makes the stock look absurd and is essentially meaningless, because 2025 net income of ~$92m is the first thin year of profit from a business whose earnings are growing 50%+ and whose two engines are at completely different points in their margin curves. The right lenses are forward EV/EBITDA, EV/gross-profit, and a sum-of-the-parts — and the single most important adjustment is the enterprise value itself.

Enterprise value, done right. AUTO1's ~$1.8bn of balance-sheet debt is non-recourse ABS funding tied to inventory and the loan book — working-capital-like, not corporate leverage [10]. With ~$745m of cash and zero corporate debt [16], the corporate enterprise value is roughly the $6.1bn market cap minus net cash ≈ $5.4bn. Treating the ABS as debt would overstate EV by a third and is the most common way to misvalue this stock.

Sources: multiples derived from a $28.0 share price (June 2026), 218.8m shares, ~$745m net cash [16], FY2025 adjusted EBITDA $233m / gross profit $1,164m [1], FY2026 guidance $287–315m EBITDA [8], and the group milestone targets [7]; consensus EPS from analyst estimates, as reported.

The sum-of-the-parts is the clarifying exercise

Because the segments are so different, the cleanest underwriting is to value them separately. The table below is illustrative, not a price target — it shows how the value splits and where the burden of proof sits.

Sources: illustrative author estimates built on segment adjusted EBITDA [3] [4], segment revenue and finance economics [13], and net cash [16]; multiples are illustrative, not company-provided.

The SOTP lands in a band that straddles the current ~$6.1bn market cap — which is the precise point. On a defensible 12–16× multiple, the profitable Merchant engine alone is worth most of today's market value; the market is paying roughly fair value for the wholesale business and pricing Autohero and the finance book as cheap-to-free optionality. If you believe the Retail loss closes and the finance book scales, the stock is undervalued; if you think the −$482 per-car loss is structural, you are paying full price for the wholesale business with execution risk attached. The bet is binary on Autohero's path to profit — which is exactly what the $5.4bn-EV-to-$825m-milestone-EBITDA multiple of ~6.5× is telling you: cheap if the milestone arrives, ordinary until it does. Sell-side sentiment leans constructive — a mean target of ~$36 versus the ~$28 price, with 11 of 14 analysts at buy or better — but the dispersion (low $23, high $44) reflects exactly this binary.

7. Management's promises versus its results

A thin-margin operator with grand long-term targets earns trust only by hitting near-term numbers. AUTO1's recent record is good — it has beaten or upgraded guidance through 2025 and is raising its tone into 2026 — which is the main reason to give the long-term milestone any credence.

Sources: Q4/FY2025 earnings call — 2025 results, 2026 guidance, and long-term 5–9% margin / 10% share targets [1] [2] [8]; Q1 FY2026 call — record quarter (249k units, $69m EBITDA, 2.5% margin) and "targeting top end" [14] [15] [16].

Q1 2026 reinforced the pattern: a record 249,000 units (+22%), $69m of adjusted EBITDA, a 2.5% margin, and Autohero growth accelerating to 48% — with management saying it may beat the top end of its full-year unit guidance [14] [15]. The 2026 guide also reveals management's deliberate choice: hold GPU broadly flat and spend the upside on volume and share, accepting slower margin expansion now to build the installed base [8]. That is a defensible land-grab strategy in a fragmented market — but it means margin progress will be lumpy, and a buyer is underwriting management's discipline as much as the model.

8. The investor's bottom line

What an intelligent investor should hold in mind simultaneously:

The structural case is strong and the runway is decades long — 3.1% share of an $800bn market, a defensible integrated model, no new entrants, and a flywheel that demonstrably lowers cost-per-car as volume grows [1] [23].

The valuation requires the future, not the present. At ~68× trailing earnings and ~24× trailing EBITDA the stock is expensive on today's numbers; at ~18× forward EBITDA and ~6.5× the long-term milestone it is reasonable-to-cheap — conditional on Autohero turning profitable and the finance book scaling without credit accidents.

The decisive variables to monitor: Retail (Autohero) adjusted EBITDA per unit crossing zero; group adjusted-EBITDA margin marching toward the 5–9% target; captive-finance attach rates and loss rates; and used-car price stability, since a principal trader's GPU and inventory are levered to price, not just volume.

Underwrite AUTO1 as a forward-EBITDA / sum-of-the-parts compounder-in-the-making, not a P/E stock — and size the position to the fact that the thesis is, at its core, a bet on management executing a still-unproven margin trajectory in a business that punishes execution errors more than almost any other. </content>

---

Figures converted from EUR at historical FX rates — see data/company.json.fx_rates for the rate table (period-end for actuals, latest rate for forward targets and enterprise value). Ratios, margins, and multiples are unitless and unchanged.

Long-Term Thesis — What Has To Be True Over 5–10 Years

The underwriting question. AUTO1 is a principal car-trading platform that has just crossed into profit. To be a superior investment over the next 5–10 years, it does not need to become a high-margin software company — that is the wrong frame. It needs to do three things in sequence and hold them: (1) keep consolidating a $803bn, structurally fragmented, barely-digitised European market through its already-profitable wholesale engine; (2) finish turning Autohero from a six-year cash sink into a real second profit engine; and (3) scale a captive-finance book into a third, higher-margin engine without a credit accident or a securitisation-funding shock. If all three happen, FY2025's $233m of adjusted EBITDA compounds toward the company's own ~$826m+ milestone and the stock re-rates off a base that is already structurally profitable. If the second and third legs stall, the durable downside is a low-single-digit-margin, cyclically price-exposed, ABS-funded car trader — which is a far less interesting asset than today's multiple assumes.

This tab is the durable frame, not a quarterly preview. It separates the part of the thesis that is already proven (the Merchant engine and the moat that survived a downturn) from the part that is still optionality (Autohero's last mile, captive finance, the 5–9% margin and 10% share ambitions that remain 2.4% and 3.1% five years after the IPO).

Sources: 3.1% share against a 10% long-term target [1]; Merchant adj. EBITDA $281.1m and group adj. EBITDA $232.1m [2] [5]; ~$826m milestone derived from the Group Milestone Targets (1,200k Merchant units at $459+/unit plus 300k Retail units at $917+/unit) [3].

1. The thesis spine — the five conditions that must hold

A long-term position in AUTO1 is a bet on five linked propositions. The first is already true; the value is in whether the other four become true. I rate each by whether the evidence today supports it, and name the single number that confirms or breaks it.

Sources: share, margin and 5–9%/10% targets [1]; FY2025 gross profit +37% and adj. EBITDA +81% [5]; Merchant 3.7% margin and downturn survival [2] [7]; Retail per-unit economics [13]; captive-finance economics [18]; self-funding and cash flow [6].

The structure of the bet is unusual and worth stating plainly: condition 1 alone roughly supports today's enterprise value — a profitable, share-gaining, #1-in-Europe wholesale marketplace is most of what you are paying for. Conditions 2–5 are the call options. That is why this is neither a value trap (there is a real, profitable floor) nor a cheap stock (you are paying for the options). The 5–10-year return is decided almost entirely by how many of the four options come into the money.

2. The compounding ladder — where $233m goes if the thesis works

The clearest way to frame the durable upside is the arithmetic of management's own milestone. The Group Milestone Targets imply 1.2m Merchant units at $459+ of adjusted EBITDA per unit (≈$550m) plus 300k Retail units at $917+ per unit (≈$275m) — roughly $826m of group adjusted EBITDA, before any standalone contribution from the finance book [3]. That is ~3.6x FY2025's $233m on ~1.8x the units — the margin doubling is the whole point, and it is what condition 2 (operating leverage) and condition 3 (Autohero crossing zero) deliver in combination.

Sources: FY2024–25 adjusted EBITDA $113m and $232.1m [5]; FY2026 guidance $287–315m (midpoint shown) [28]; LT milestone derived from the Group Milestone Targets [3]; FY2023 adjusted EBITDA derived from segment disclosures.

The milestone is undated, which is the honest caveat: management frames Merchant at a 10–15% annual growth corridor and Retail at 20–40% [3], implying the $826m+ is a 7–10-year, not a 3-year, destination. At a 7-year horizon that is a ~20% EBITDA CAGR; at 10 years, ~14%. The durable thesis is therefore a low-teens-to-twenties EBITDA compounder off a profitable base — not a hyper-growth story and not a deep-value one. The PM's job is to underwrite the slope of that line, not any single quarter on it.

3. The durable engine — a decades-long consolidation runway

The reason to believe the slope is real is structural, and it is the most durable fact in the file. AUTO1 sits in a $803bn used-car market (plus a ~$115bn financing pool) [9] that is extraordinarily fragmented — more than 250,000 dealers, with the top 20 owning under 6% [8]. A pan-European platform that can source, price, refurbish and move metal cheaper than a long tail of sub-scale lots is built to consolidate that tail, one car at a time, for a very long time.

The runway is quantified, not asserted. On the supply side, a 190m-vehicle car parc yields ~10–15m units a year of addressable consumer-to-business selling demand, of which AUTO1 targets capturing 2.5–3.0m long-term [10]. On the demand side, it frames ~10m units of external dealer-sourcing demand and captures only 7% today, against a 20–25% long-term ambition [11]. Whichever lens you use, penetration is low single digits — the consolidation is multi-decade, and that is exactly what makes the growth durable rather than a sprint that ends.

Sources: 3.1% group share vs 10% target [1]; 7% of external sourcing demand today vs 20–25% long-term (midpoint shown) [11].

What makes the consolidation defensible rather than merely available is the moat the durability test already validated. The advantage is unusually physical: a proprietary dataset of 6m+ transactions over 14 years, 170+ logistics centres, 12 production centres with 248k units of refurbishment capacity, and 750+ branches — built over more than a decade [14]. Management's sharpest moat claim is data exclusivity: AI pricing needs settled transaction prices, which are private and only generated by trading, whereas classifieds hold only asking prices — a car "cannot be conquered by an AI prompt" [15]. The decisive real-world evidence is not a slide: in the 2022 downturn AUTO1 stayed Europe's largest used-car dealer and kept its wholesale engine profitable while two better-funded online challengers, Cazoo and Carnext, exited most of its markets [7]. A moat that outlasts better-capitalised entrants through a trough is the kind worth underwriting for a decade.

4. The two engines, on opposite ends of their margin curves

Un-blending the segments is the only way to underwrite the next five years correctly, because the consolidated 2.4% margin averages two opposite economic profiles — and the entire long-term debate lives in the gap between them.

Sources: Merchant segment adj. EBITDA $70.9m (FY2021) to $281.1m (FY2025) [2]; Retail segment adj. EBITDA −$192.2m (FY2021) to −$48.9m (FY2025) [4].

Merchant is the proven compounder. It earns $380 of adjusted EBITDA per car today on $1,147 of GPU, with a long-term target of $459+ per unit on $1,175+ GPU [12]. Its margin is still expanding (3.1% to 3.7% in one year) as scale fills the fixed network. On its own it would screen as a decent, defensible marketplace — and it already earns more than the entire group's adjusted EBITDA, meaning every dollar of Retail loss is a drag the market can see being removed.

Retail (Autohero) is the swing factor. It carries a far larger GPU ($3,100, target $3,784+) against a far heavier per-car cost base, so it still loses $482 per car — but that loss has compressed from −$4,818 in FY2021, and management states FY2025 retail unit economics were already positive before marketing [13]. The long-term target is $917+ of EBITDA per unit on 300k cars — the ~$275m swing that turns the group from a single-engine into a twin-engine compounder. This is the highest-leverage single variable in the whole thesis, and it is also the one the multiple is most exposed to: the stock fell ~20% intraday on results day when 2026 guidance implied the margin pause would persist.

The honest watch-item on Retail quality: reported GPU is flattered because in-house refurbishment is capitalised into inventory rather than expensed — a swing UBS pressed management on at ~$83m in Q4 2025 alone [30]. It is disclosed and analyst-tested, not a red flag — but the gap between reported GPU and cash margin is the metric to track as the segment approaches break-even.

5. The third engine almost nobody underwrites — captive finance

The least-appreciated long-term driver is the embedded lending layer, where the highest-quality incremental margin lives because interest income carries far better economics than trading metal. On the consumer side, captive internal interest already contributes $247 per car of Autohero GPU (up from $14 in FY2021), with a stated long-term captive contribution of ~$998 per car [17]. The disclosed unit economics are genuinely attractive for a lender — net interest margin ~5% (target 5–7%), attachment 40% (target 50–60%) [18].

Source: Capital Markets Event 2026 — Retail Finance GPU, captive internal interest per car, with a ~$998 long-term outlook [17].

The strategic logic is sound — financing turns a one-off transaction into a multi-year relationship and lifts switching cost. But this is the engine that also creates the most dangerous failure mode, and the two cannot be separated. A loan book consumes equity (AUTO1 retains the first-loss tranche of its ABS), it brings consumer-credit conduct and AML regulation [31], and it invites credit losses: management already booked a $14m merchant-finance impairment in 2025 from a single new-market underwriting lapse [19]. A young, fast-growing book is exactly where under-reserving hides. Underwrite captive finance as optionality with a tail risk, not as a clean margin gift — a second credit charge would signal the lending build is outrunning underwriting discipline, and the CFO change to a consumer-credit specialist (Wallentin, ex-Hoist Finance) is the tell that the finance arm is becoming central, not incidental.

6. The reinvestment runway and the self-funding question — the thesis breaker

The durable-thesis question that matters most is not growth — it is how the growth is funded, and whether that funding survives a credit cycle. Here the long-term frame must be precise, because the same fact reads as a strength or a fatal flaw depending on the next stress.

On an IFRS basis AUTO1 burns cash: FY2025 operating cash flow was −$544m and free cash flow −$575m. Yet it ended the year with ~$705m of cash and no corporate debt [6]. Both are true. The bridge is non-recourse asset-backed securitisation: AUTO1 funds inventory and the finance book through ABS at ~80%+ advance rates, which IFRS records as financing inflows while the asset build lands as an operating outflow. Management's preferred "AUTO1 Cash Flow" — which nets the matched ABS funding against the asset growth — was positive $76m in FY2025.

Source: Capital Markets Event 2026 — AUTO1 Cash Flow walk for FY2025 (adj. EBITDA $233m to +$76m after net inventory −$46m and net captive finance −$76m post-ABS) [23]. Intermediate items grouped.

The honest long-term read: the trading business is genuinely self-funding (inventory is overwhelmingly ABS-covered, capex is ~25bps of revenue), and management has committed to a clear funding framework that maintains a direct link between the asset base and its re-financing while delivering positive AUTO1 Cash Flow [22]. The real equity call is the captive-finance build, which outran its ABS funding by ~$76m in FY2025. So "self-funding" is fair for the car business and optimistic for the finance build — which is precisely why the ~$705m cash cushion and zero corporate debt matter as durability insurance.

But this is also where the single most dangerous failure mode sits, and it is structural rather than operational. AUTO1's own FY2022 report names liquidity "the most relevant potential financial risk" and concedes it will need continued access to banks and capital markets until it reaches positive operating cash flow [21]. With both the inventory ABS and the consumer-finance ABS roughly 87% drawn at Q1 2026 [20], the model is hostage to continuous, cheap securitisation access. A credit-market shock that widens ABS spreads or cuts advance rates would hit the model exactly where the income statement does not show it — and no amount of EBITDA growth offsets a funding-market that closes. The durable thesis-breaker is not Autohero stalling; it is operating cash flow that never turns positive without fresh ABS draws.

7. Capital allocation and management — the multi-year track record

A thin-margin operator with grand long-term targets earns the right to be underwritten for a decade only through alignment and a demonstrated track record. AUTO1's is good on near-term execution and unproven on the grand promises — the exact split a long-term investor must price.

Alignment is genuine and rare. This is a founder-run company since 2012; the two founders still hold ~21% of the equity, and CEO Bertermann drew $544k of fixed cash with zero variable pay, dwarfed three orders of magnitude by his equity stake [26]. Management runs a debt-free, self-funding balance sheet rather than tapping shareholders with dilutive cash calls — removing the most common minority-holder fear at a high-growth platform. The credit-side discipline is the capital-allocation story: reinvestment flows into the asset base (inventory, the finance book, the network) funded by ABS, not equity.

The promise-vs-delivery record splits cleanly into two buckets. Near-term, self-set guidance has been kept or beaten repeatedly — group EBITDA breakeven delivered a quarter early after the decisive Q2 2022 profit pivot [27], Merchant GPU guidance raised and beaten through the cycle, full-year EBITDA guidance raised and beaten in both 2024 and 2025. The grand IPO promises remain open five years on.

Sources: early breakeven after the Q2 2022 pivot [27]; FY2025 result and 2026 guidance [5]; long-term margin and share targets [1]; Retail segment loss [4].

The IPO context sharpens the underwriting. AUTO1 listed in February 2021 at $43.04 (~$8.95bn) on a pitch to make Autohero "the leading retailer for used cars in Europe" with ~$849m of growth investment [24] [25]. The shares now trade near $28 — well below IPO — so even the aligned founders have not delivered for IPO-era buyers, and the very same ~$680bn-market / 5%-CAGR / offline-to-online thesis sold at listing is still substantially un-penetrated [24]. The lesson for a 5–10-year holder: trust this team on the next quarter; verify them on the next five years. Their execution credibility is high and rising; their decade-defining promises are still pending their verdict.

8. What proves the thesis is working — or breaking

The point of the durable frame is to give a PM a scorecard that separates long-term thesis evidence from quarterly noise. Each signal below is the leading indicator of one of the five spine conditions.

Sources: Merchant margin [2]; Retail per-unit economics [13]; captive credit loss [19]; cash flow and ABS draw levels [6] [20]; entrant view [16].

The most recent data point cuts in the thesis's favour on the operating side: Q1 2026 was a record 249,000 units (+22%), $331m gross profit and $69m adjusted EBITDA, with management saying it may beat the top end of full-year unit guidance [29]. But the same quarter's $69m was only ~$2m above the prior year — the margin pause is visible — and management currently reports no new entrants attacking the sourcing or retail side [16]. Watch the signals, not the headline records.

9. Valuation over the horizon — what the market is paying for

The headline P/E is a trap (~68x trailing, ~39x FY2026E) because FY2025 is the first thin year of profit. The right long-term lens is forward EV/EBITDA against the milestone, with the enterprise value corrected for the ABS confusion: the ~$1.88bn of balance-sheet debt is non-recourse, working-capital-like ABS, so with ~$705m cash and zero corporate debt the corporate EV is roughly the ~$6.08bn market cap less net cash ≈ $5.39bn [6]. Treating the ABS as corporate debt overstates EV by a third and is the most common way to misvalue the stock.

On that corrected EV, the multiple tells the whole durable story in one number: ~6.5x the ~$826m+ long-term milestone EBITDA [3]. That is cheap if the milestone arrives and ordinary until it does — which is exactly the binary an intelligent investor is underwriting. The market is paying roughly fair value for the proven Merchant engine and pricing Autohero and the finance book as cheap-to-free optionality. If you believe conditions 3–5 come into the money over 5–10 years, the stock is materially undervalued; if you think the Retail loss and the funding dependence are structural, you are paying full price for a cyclical car trader with execution risk attached.

---

Competition — Who Can Hurt AUTO1, Who It Can Beat

Figures converted from EUR at historical FX rates — see data/company.json.fx_rates for the rate table. Ratios, margins, and multiples are unitless and unchanged. Auto Trader figures remain in GBP, as no GBP/USD rate is available in this corpus.

AUTO1 is a three-engine business — AUTO1.com (B2B wholesale, its profit centre), Autohero (B2C online retail, its growth engine) and wirkaufendeinauto.de (C2B consumer sourcing). The competitive question is whether the moat under those engines is real. The short answer: the wholesale moat is real and widening; the retail engine is winning share but still loses money, and that is where the company is most exposed.

Bottom line — a real, widening wholesale moat wrapped around a still-unproven retail bet

AUTO1 describes itself as the largest used-car dealer in Europe, operating in a highly competitive sector whose main rivals are independent used-car dealers, small-ads websites and apps, rental-car fleets and professional dealers — and it notes that two venture-funded online challengers, Cazoo and Carnext, exited most of its markets in 2022 [1]. That tells you the shape of the arena: a vast, hyper-fragmented market — over 250,000 used-car dealers in Europe, the top 20 of whom own less than 6% of it [2] — in which AUTO1 is consolidating share rather than fighting a single named giant.

The verdict. AUTO1's advantage is real and structural in wholesale (AUTO1.com) and real-but-unfinished in retail (Autohero). Three assets make the wholesale moat genuine: a proprietary pricing dataset of 6M+ transactions accumulated over 14 years, the largest pan-European used-car logistics network (170+ logistics compounds, 1.5m transports a year), and an ABS-funded balance sheet — assets management argues "cannot be conquered by an AI prompt" [3]. Share is gaining: AUTO1 reached a record 3.1% European share in 2025, growing roughly 14x the rate of the overall used-car market [4].

The single competitor type that matters most is not a company — it is the market cycle acting on the retail ramp. Management sees no new competitive entrants and calls itself "our own competition" [5]; the genuine 24-month threat is a softening European used-car market — already visible in France, which French peer Aramis flagged as weak on the same morning as AUTO1's Q1 2026 call [6] — stalling Autohero's margin climb while it is still loss-making. Among named rivals, Carvana is the one that matters: not because it competes in Europe today, but because it is the live proof that the vertically integrated online-retail model can reach strong profitability — and the template a deep-pocketed entrant would copy.

Sources: revenue and units — Capital Markets Event, June 2026, segment financials [7]; 3.1% European share [4]; market cap derived from 218.8m shares at the $27.98 (€24.40) close (19 Jun 2026), company price feed, as reported.

The arena and the peer set — why these five-to-six

The market AUTO1 fights in is an $823bn European used-car market (plus ~$118bn of financing) [8] that is stable in volume but extraordinarily fragmented [2]. No listed pure-play maps onto all three of AUTO1's engines, so the right peer set is built by engine, and each peer below was confirmed from its own filing to run the model claimed:

• Aramis Group (ARAMI) — the closest direct peer. A European B2C online used-car retailer that buys, refurbishes and sells to private individuals and dealers, selling 150,074 used vehicles (119,109 of them B2C) in its year to September 2025 [9] across nine industrial refurbishing centres [10] — the same buy-refurbish-retail model as Autohero, in the same region.
• Carvana (CVNA) — the global benchmark. Self-described "leading e-commerce platform for buying and selling used cars," combining online retail with a vertically integrated supply chain [11]. Different continent, same model — and the one that has proven the economics.
• OPENLANE (OPLN) — a B2B read-across to AUTO1.com. A "leading digital marketplace for wholesale used vehicles" across the US, Canada and Europe, with ~1.5m annual transactions and $28.8bn of GMV in 2025 [12]; its European cross-border remarketing competes head-on with AUTO1's wholesale platform.
• ACV Auctions (ACVA) — a pure digital B2B wholesale "marketplace platform for wholesale vehicle transactions and data services" connecting dealers [13] — a functional comp to AUTO1.com, in the US.
• CarMax (KMX) — the integrated-retail benchmark at scale. The largest US used-vehicle retailer (780,684 retail units) and a major wholesale auctioneer (538,203 units) in the year to February 2026 [14] — what Autohero is trying to become, with a wholesale read-across too.
• Auto Trader (AUTO) — kept as an adjacency, not a like-for-like. The UK's largest automotive marketplace [15], but it runs an advertising/classifieds model and does not take cars onto its balance sheet the way AUTO1 does — useful for digital-demand economics, not for unit economics.

The genuinely relevant private rivals named around AUTO1 — Cars24, AutoScout24 and the now-collapsed Cazoo/Carnext — have no listed filing to benchmark and are flagged here for completeness rather than tabled.

Peer comparison

Revenue/net income from staged financial feed (latest FY), EUR figures converted to USD at the FY2025 period-end rate (1.175); Auto Trader remains in GBP (no GBP/USD rate in corpus). AUTO1 corroborated by the segment financials in the June 2026 Capital Markets Event [7]. Business-model confirmations cited per peer above: ARAMI [9], CVNA [11], OPLN [12], KMX [14], ACVA [13], AUTO [15]. Auto Trader's margin/ROCE are not comparable (advertising model). Market cap/EV unavailable for peers — see valuation table below.

The table makes the moat debate concrete. AUTO1 is by far the largest by revenue after CarMax, yet its net margin (0.95%) is the thinnest of any profitable peer — a direct consequence of the principal buy/sell model (revenue is the full car price) and of carrying a loss-making retail engine. The capital-light wholesale platforms (OPENLANE 9.2% net margin, Auto Trader 47%) and the proven online retailer (Carvana 6.9%) all out-earn AUTO1 per unit of revenue today. That is the gap AUTO1's long-term targets are built to close.

Net margin from staged financials (net income ÷ revenue); ROCE from staged ratio feed. AUTO1 figures corroborated by Capital Markets Event segment financials [7]; CarMax ROCE not available in feed; ACV and Auto Trader omitted (negative / non-comparable advertising economics).

Coverage of every named public competitor — market cap and enterprise value

The staged price feed failed for all six peers, so market cap and enterprise value are not reliably available in this corpus for them and are marked N/A rather than invented. AUTO1's own market cap is derived from its share count and last close in the price feed.

AUTO1 market cap/EV derived from the company price feed (share count x last close) and staged balance-sheet net debt, converted to USD at the 1.1467 rate (21 Jun 2026), as reported; peer market cap/EV genuinely absent from the staged feed and corpus — marked N/A with reason, not estimated.

Where AUTO1 wins

1. The wholesale platform is structurally #1 in Europe — a real network effect. AUTO1.com is Europe's #1 wholesale platform for used cars, with ~54,000 unique buying dealers and 50,000+ cars listed across 30+ countries [16]. Pan-European supply aggregated against pan-European demand produces prices a single local market cannot — the classic two-sided flywheel. No European listed pure-play wholesale competitor of comparable scale exists; the nearest functional comps (ACV, OPENLANE) operate primarily in North America.

2. A proprietary data + logistics moat that compounds with every trade. AUTO1 owns "the largest and most comprehensive pricing data set for the European used-car market," built from 6M+ transactions over 14 years, and argues its AI pricing models "cannot be replicated without being us or going through the same history of trades" [3]. Crucially, data is paired with the largest used-car logistics network in Europe — 1.5m transports a year across 170+ compounds [17] — the physical layer that "cannot be conquered by an AI prompt" [3]. Its own filings frame this network — 400+ delivery points, up to 130 warehouses — as a market-entry barrier for potential competitors [18].

3. Share is being taken — and faster than anyone else. AUTO1 grew European share to 3.1% in 2025 (+50bps), ~14x the rate of the overall market [4], and management states plainly that "no other public auto retailer in the EU grew faster last year than us" [17]. The Merchant network keeps widening — 36,200 active buying partners in Q1 2026, up 23% year-on-year, with the company deliberately prioritising dealer share over basket size [19].

4. Autohero is the fastest-growing car retailer in Europe. The retail brand is positioned as Europe's fastest-growing car retailer [20], compounding at a 58.5% CAGR since inception with FY2025 GPU of $3,061, versus the Merchant engine's 51.7% CAGR and $1,147 GPU [21]. Against its closest listed peer Aramis (150,074 units to AUTO1's ~101,500 Autohero units but across a far larger refurbishing base), Autohero is the faster grower.

2025 share of 3.1% and the +50bps step (implying ~2.6% in 2024) from the Q4 FY2025 call [4]; 10% long-term target from the June 2026 Capital Markets Event [2].

Where competitors are better

1. Carvana has proven the profitability AUTO1 has not. Carvana earned a ~6.9% net margin and 16.1% ROCE in FY2025 on $20.3bn of revenue — the highest returns in the peer set among balance-sheet retailers — demonstrating that the vertically integrated online model [11] can throw off real profit at scale. AUTO1's Autohero engine still runs a negative adjusted EBITDA margin (-2.4% in FY2025); Carvana is the benchmark proving the destination is reachable — and the model a US entrant could bring to Europe.

2. The capital-light wholesale platforms out-earn AUTO1 per unit. OPENLANE — a direct B2B comp with European exposure — converts $28.8bn of GMV into a 9.2% net margin without taking cars onto its balance sheet [12]. AUTO1's principal model carries inventory and books the full car price as revenue, structurally compressing margin to under 1%. In pure capital efficiency, the marketplace model wins — a reminder that AUTO1's scale is balance-sheet-heavy.

3. CarMax shows the scale ceiling — and the auction read-across. CarMax retails 780,684 units and auctions another 538,203 a year [14] — roughly 8x Autohero's retail volume — at a scaled, profitable integrated operation. It is what AUTO1 aspires to, evidence the integrated model endures, and a competitor whose wholesale auction business read across to AUTO1.com's economics.

4. Aramis runs a larger industrial refurbishing base in AUTO1's own backyard. Aramis operates nine refurbishing centres and sold 150,074 used vehicles (119,109 B2C) in Europe [9] [10] — more B2C retail units than Autohero today, and a direct competitor for the same European sellers and buyers, particularly in France where it leads.

Segment divergence — the heart of the competitive question

Segment revenue from the June 2026 Capital Markets Event: Merchant financials [7] and Retail financials [22]; EUR converted to USD at each fiscal year's period-end rate.

Merchant adjusted EBITDA margin (1.5% to 3.7%) from the Capital Markets Event Merchant financials [23]; Retail adjusted EBITDA margin (-29.3% to -2.4%) from the Retail financials [22]. Margins are unitless and unchanged from the native-currency view.

The two charts together are the thesis. The Merchant engine is a scaled, profitable, share-gaining wholesale platform whose margin has more than doubled to 3.7%; the Retail engine is a fast-growing but still loss-making retailer narrowing from -29% to -2.4%. The competitive risk is concentrated entirely in the second chart's last mile to break-even — and management itself flags that faster retail growth is a near-term headwind to unit economics (roughly $120 of EBITDA-per-unit drag for every extra 10 points of growth) [24].

Threat assessment

Threat evidence: France/cycle softness [6] and contained-macro view [25]; Carvana model [11]; OPENLANE/ACV wholesale platforms [12] [13]; private challengers, classifieds and e-commerce/OEM entrants named in AUTO1 risk factors [1]; AI/no-entrant view [5].

The threat map underlines the bottom line: the dangerous box is the top row — cyclical, not competitive. Management's own read is that the macro impact in early 2026 was "limited and contained," even as France showed a strong negative pull in April [25]. Direct competitive entry is, on the present evidence, a low-probability event AUTO1's data-plus-logistics moat is built to absorb.

Moat watchpoints — what would change the call

These are the few signals that would actually move the competitive verdict, with the disclosed anchor for each:

Anchors: 3.1% share [4]; Retail adjusted EBITDA margin -2.4% [22]; Merchant GPU $1,147 and $1,204+ milestone target [26]; brand awareness 35% [27]; cash flow and 83% inventory ABS loan-to-value [28].

The watchlist resolves to one question: does Autohero reach break-even before the used-car cycle tests it? The wholesale moat is not in doubt; the retail ramp is. An investor who sees share keep climbing toward 10%, Retail EBITDA cross zero, and Merchant GPU march to its $1,204+ target [26] is watching a moat widen. The reverse — flat GPU, stuck Retail losses, rising marketing spend — would mark commoditisation setting in.

---

Current Setup & Catalysts — Where We Are Now

Figures converted from EUR at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

The one-line read. AUTO1 has done the hard operating work — first full-year net profit, record adjusted EBITDA, a record Q1 — and the stock has already re-rated for it, round-tripping from a ~$16.58 spring low to $27.98 on a +62% three-month rally into the 17 June Capital Markets Event. So the operating setup is good and the price setup is no longer cheap: a crowded-long consensus is paying ~39x FY2026 EPS for a swing factor — Autohero's margin inflection — that management has explicitly pushed into 2027. The single most decision-relevant event is the next print (Q2/H1 2026, a late-July/early-August window): not the unit headline, which is already strong, but Autohero's segment-level adjusted EBITDA per unit and group GPU — the exact variables the multiple is leveraged to. Everything else on the calendar is information; that print is resolution.

This page is the bridge between the durable 5-to-10-year thesis and the near-term evidence path — not a news digest. AUTO1 is not a binary or distressed name: condition 1 of the long-term thesis (a profitable, share-gaining Merchant engine) is already proven and roughly supports today's enterprise value. So no single quarter decides the whole case. What the near-term calendar can do is update the call options — Autohero crossing zero, captive finance scaling cleanly, the ABS-funded model holding — that the rest of the value rides on.

Source: price, returns and 52-week range from the staged price feed (data/tech/levels.json, momentum.json); 52-week range $16.58–$35.57. Next-results date is an estimated window from the prior-year cadence (30 Jul 2025, 31 Jul 2024), not an officially confirmed date.

The variant view, sized

The web work is right that the bull case here is the consensus — 11 Buys against 2 Holds and 1 Sell, a mean target of $35.43 (~27% upside). The edge is therefore not in out-arguing the direction; it is in the timing and quality of the margin step the Street has already underwritten for 2027.

• Where I sit vs the Street. Consensus models FY2027 group adjusted EBITDA of roughly $438m (≈$1.09 EPS) — a ~46% step from the FY2026 guidance midpoint of $300m that requires Autohero to swing from a −$49m drag toward breakeven and Merchant to keep compounding. Management itself has framed FY2026 as a "hold GPU broadly flat, buy volume" year, with adjusted-EBITDA growth coming from units rather than margin [3]. That pushes the entire re-rating burden onto a 2027 inflection that has slipped before. I model FY2027 nearer $390–407m if the inflection lands late — ~8–11% below the Street, and that gap is what re-rates or de-rates the multiple.
• The offsetting tailwind (why this is not a short). Q1 2026 ran hot — a record 249,000 units (+22%), Retail units +48%, $331m gross profit, $69m adjusted EBITDA — and management is now "targeting the top end" of FY2026 guidance, noting Autohero volumes may exceed the range [6] [7]. The near-term risk is margin/mix, not volume.

The honest conclusion: I am directionally aligned with consensus on the proven Merchant engine but below the Street on the pace of the 2027 margin step, and into a name where +62% of good news is already banked. That asymmetry — limited contrarian cushion on the upside, an outsized down-move if margin disappoints — is the setup, and it is exactly what the Feb-2026 tape already demonstrated.

Source: FY2025 actual $232m and FY2026 guidance $287–315m (midpoint shown) per company filings [1] [2]; FY2027 consensus per the analyst-estimates feed (data/estimates/); the FY2027 "our model" figure is the analyst's variant estimate.

What changed in the last 3–6 months

The recent setup is best read as a violent round-trip around one debate — margin — bracketed by a record operating cadence. Three events define it, and the tape's reaction to each is the most useful thing on the page.

Sources: event facts cross-read to the FY2025 results deck [1], the Q1 2026 call [6], and the Capital Markets Event milestone targets [12]; price reactions per the indexed news file [17] and the staged price feed.

The narrative arc. What investors worried about before (early 2026) was whether the FY2025 profit was real and bankable; the record print answered that. What they worry about now is narrower and harder: the FY2026 "volume-over-margin" choice means the GPU-flat year is a deliberate pause, so the question has shifted to whether the 2027 margin inflection actually arrives — and, underneath it, whether a model that posts record adjusted EBITDA while burning IFRS cash and growing a captive loan book ~50% can keep funding itself through the ABS market. Two things are not contested: there is no short-seller report, probe, auditor flag or restatement in the public record, and the founders (~21%) plus a debt-free, self-funding balance sheet remove the usual dilution fear.

The historical price-reaction base rate

Any "high impact" claim on the next print has to be anchored in how this stock actually moves on events — and AUTO1's pattern is unusually clear and unusually asymmetric. Clean consensus-surprise history is thin (the name reports on a German half-yearly cadence with limited per-quarter EPS-surprise data), so the better gauge is the realized move around each recent catalyst.

Source: realized moves from the staged price feed (data/tech/, data/prices/) and the indexed news file [17]; 52-week range $16.58–$35.57.

The base-rate takeaway, in numbers: AUTO1's events produce outsized moves — the 52-week range is a 2.1x spread ($16.58 to $35.57), and the single largest reaction (the Feb guide) was roughly −20% in a day even though the headline FY2025 EBITDA beat. The asymmetry is the signal: margin/guidance disappointments are punished far harder than unit beats are rewarded. That is why a margin-quality miss on the next print is a genuine high-impact, double-digit-move risk, not a vibe — and why the +62% rally now banked into a crowded long removes most of the cushion on the upside.

The live debate — what the market is watching now

Sources: cash-flow framing and the +$76m AUTO1 Cash Flow [13]; the $13.9m merchant-finance credit charge [9]; Merchant 3.7% margin and per-unit economics [11] [10]; valuation/consensus per the analyst-estimates feed.

Ranked catalyst timeline

The table below is the core deliverable: the best near-term catalysts ranked by decision value to an institutional investor, not by date. AUTO1's archetype — a profitable-but-thin-margin principal trader with an emerging captive-finance book and no public short/borrow data — drives the column set: I add a positioning column (crowding amplifies the move) and keep credit/funding rows prominent. Note the calendar is cadence-driven, not hard-dated: AUTO1's exact reporting dates are not officially confirmed in the corpus, so date confidence is Medium even where the thesis impact is High.

Sources for the dated commitments and windows in this table: FY2026 guidance $287–315m / 940k–1.0m units [2]; "GPU broadly flat" and self-funding framing [3]; adjusted-EBITDA growth to outpace unit growth [4]; "targeting the top end" and Autohero may exceed range [7]; Retail EBITDA/unit −$482 and targets [10]; $13.9m credit charge [9]; ABS facilities ~87% drawn [14]; FY2027 consensus and target/EPS revisions per the analyst-estimates feed.

Impact / decision view — what resolves the debate vs what merely informs

Not every catalyst closes the underwriting question. The Retail-margin prints and the credit/funding signals actually resolve durable thesis variables; the rest add information.

Sources: the resolving-evidence definitions track the long-term thesis spine — Retail per-unit economics [10], captive credit [9], and the self-funding/ABS framing [13] [14].

The next 90 days

The honest read on the calendar: there is no hard-dated binary event in the next 90 days. The one decision-relevant print — Q2/H1 2026 results — falls in a late-July/early-August window (estimated from the prior-year cadence of 30 Jul 2025 and 31 Jul 2024, not an officially confirmed date), and the cleaner read on the 2027 setup is the Q3 print in early November, just beyond 90 days. So the next quarter is the live event; the quarter after is the verdict.

Sources: the GPU-flat 2026 plan [3]; "no corporate debt" framing [3] and ABS draw levels [14]; next-results window from the prior-year reporting cadence.

What would change the view

Three observable signals over the next ~6 months would most change the investment debate — this is the event path that would force a thesis update, not the final verdict:

1. Autohero (Retail) segment-level adjusted EBITDA per unit. Two consecutive quarters positive (after marketing) validates the Bull's swing factor and the 5–9% group-margin path [10]; two consecutive quarters worsening while GPU stays flat confirms the Bear that the volume-led 2026 plan is diluting, not building, margin.
2. A second captive-finance credit charge. After the $13.9m FY2025 "one-off" [9], a second similar charge as receivables scale ~50% would confirm under-reserving and turn the finance leg from optionality into a liability — the Forensic/Bear lever. Clean credit as the book seasons does the opposite.
3. Group IFRS operating cash flow vs ABS draws. OCF turning positive without fresh ABS draws would prove the model self-funds and de-risk the durable thesis-breaker [13]; ABS spreads widening or advance rates cut while both facilities sit ~87% drawn [14] would hit the model exactly where the income statement does not show it — the company's own filings name liquidity "the most relevant potential financial risk" until it reaches positive operating cash flow [15].

---

Figures converted from EUR at historical FX rates - see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Bull and Bear

Verdict: Lean Long, Wait For Confirmation - the business has genuinely inflected (first net profit, real operating leverage, a scaled and moated Merchant engine), but at roughly 39x FY2026 EPS the price already pays for an Autohero turn that has not yet crossed zero. Bull and Bear are not arguing about whether AUTO1 is a good business; they agree the wholesale Merchant platform earns all $281m of group adjusted EBITDA [1]. They are arguing about the part that is not yet proven: whether the retail arm, Autohero, finishes its long glide from -$4,644 per unit to its current -$482 per unit and turns the loss center into a second profit engine [2]. The tension that matters most is therefore not the cheapness of the Merchant engine - both sides concede that - but whether the retail turn and the self-funding of growth are real, observable, and durable. The verdict tilts long because the realized operating leverage and the moated wholesale platform are a floor a pure value trap does not have, but it waits because the decisive evidence - Autohero crossing zero and operating cash flow turning positive without fresh securitization draws - has not yet landed.

Bull Case

The three sharpest points carry forward Bull's strongest, already-sourced claims. The operating-leverage inflection is realized rather than promised: group operating income swung from -$223m in FY2022 to +$167m in FY2025, delivering the first-ever full-year net profit of $92m and record adjusted EBITDA of $232m (+81%) on 842,271 units (+22%) and $1,164m of gross profit (+37%) [3] - and opex pre-SDI grew 28.9% while gross profit grew 36.7%, with management guiding adjusted-EBITDA growth to again outpace unit growth in 2026 [4]. The Merchant platform already earns more than the whole group, at $281m of adjusted EBITDA on a 3.7% margin [1], and it proved its moat in a downturn: in 2022 AUTO1 stayed the largest used-car dealer in Europe while better-funded challengers Cazoo and Carnext left most of its markets [5], reaching a record 3.1% share against a 10% long-term target [6]. And Autohero is one to two quarters from the swing: per-unit adjusted EBITDA improved from -$4,644 (FY2021) to -$482 (FY2025) with retail GPU compounding from $410 to $3,100 toward a positive $1,660-$2,760 target [2], against group adjusted-EBITDA guidance of $287-315m for FY2026 [7].

Sources: bull points sourced as cited above - FY2025 Results Trading Update [3]; FY2026 guidance [7]; 2026 Capital Markets Event [1] [2]; Q4 FY2025 call [4] [6]; FY2022 Annual Report [5].

Bull's price target is $41.3 per share (from $28.0, about +48%), by re-rating to roughly 21x FY2027e corporate EV/adjusted EBITDA as Autohero crosses per-unit break-even and group adjusted EBITDA scales to about $384-396m, cross-checked to roughly 38x FY2027e EPS of $1.09, over a 12-18 month timeline. The disconfirming signal Bull names: retail adjusted EBITDA per unit reversing for two consecutive quarters while a second captive-finance credit charge lands - evidence the swing factor is stalling and the lending build is outrunning underwriting discipline.

Bear Case

The three sharpest points carry forward Bear's strongest, already-sourced claims. First, the premium pays for Autohero: at roughly 39x FY2026 EPS ($0.72) and about 18x EV/adjusted EBITDA on the $287-315m FY2026 guide [7], the stock prices a successful retail ramp, yet Autohero lost $49m (a -2.4% margin) in FY2025 and the entire group profit is the Merchant engine [1]; the shares fell roughly 20% intraday on 25 Feb 2026 when the guide implied stalling margin. Second, the $92m net profit cannot be banked: FY2025 operating cash flow was -$544m and free cash flow -$574m, with the cash position held only because about $560m of financing draws refilled the burn - structural every year since FY2022, a year in which AUTO1 itself named liquidity as the most relevant financial risk and conceded it would need access to banks and capital markets until it reached positive operating cash flow [8]; the inventory ABS facility was 87% drawn and the consumer-finance facility also 87% drawn at Q1 2026 [9], and management calls this "no corporate debt" while running the model on continuously rolling non-recourse securitizations [10]. Third, AUTO1 is increasingly a balance-sheet lender carried at a platform multiple: captive-finance receivables grew about 50% in 2025, a $13.9m merchant-credit loss was booked and framed as a one-off on a book still scaling fast, and the consumer-finance ABS sits at 87% drawn [9] - a young, fast-growing book is exactly where under-reserving hides, and the credit cycle has not been tested.

Sources: bear points sourced as cited above - FY2026 guidance [7]; 2026 Capital Markets Event [1]; FY2022 Annual Report, Liquidity Risk [8]; Q1 FY2026 Results Presentation [9]; Q4 FY2025 call [10]. The $13.9m credit loss is carried forward from the forensic review as a flag and is not pinned to a corpus page.

Bear's downside target is $16.6 per share (about 40% below the $28.0 close), by compressing EV/adjusted EBITDA from about 18x to 10-11x on the FY2026 midpoint (about $300m) - valuing AUTO1 as a cyclically exposed, ABS-funded principal car trader with no Autohero re-rating plus a funding/credit-risk discount, adding back about $0.7bn corporate net cash, cross-checked to roughly 15x FY2027 EPS - over a 12-18 month timeline. The cover signal Bear names: two consecutive quarters of Autohero positive segment-level adjusted EBITDA while group IFRS operating cash flow turns positive without new ABS draws.

The Real Debate

Each row below compares the two sides on the same underlying fact. The shared facts are anchored to the primary record: the cash bridge and ABS draw levels are cited from the Q4 FY2025 call and Q1 FY2026 presentation [10] [9], the Autohero per-unit figure from the 2026 Capital Markets Event [2], and the FY2026 guide from the trading update [7].

Sources: shared facts traced to the Q4 FY2025 call [10], the Q1 FY2026 presentation [9], the 2026 Capital Markets Event [2], and the FY2025 trading update [7].

Verdict

Lean Long, Wait For Confirmation. Bull carries more weight on what is already proven: the Merchant engine is a scaled, share-gaining, structurally profitable platform that earns all $281m of group adjusted EBITDA and out-survived the downturn that killed Cazoo and Carnext [1] [5], and the FY2025 swing to a $92m net profit on rising operating leverage is a real floor that a value trap does not have. The single most important tension is whether Autohero crosses zero: at -$482 per unit the glide path is genuinely close [2], but Bear is right that the price already pays for that turn and that the IPO promises of a 5-9% margin and 10% share are 2.4% and 3.1% five years on, so the multiple has no cushion - a fact the 20% guidance-day drop demonstrated. Bear could still be right if the European used-car cycle softens into the last mile or if the captive-finance book takes a second credit charge that exposes chronic under-reserving. The condition that flips this to a clear long is the confirmation itself: two consecutive quarters of positive segment-level Autohero adjusted EBITDA. Separating the markers, the durable thesis breaker is the funding model - operating cash flow that never turns positive without fresh ABS draws, or a second credit charge confirming under-reserving, which would invalidate the self-funding case entirely [10]; the near-term evidence marker is simply Autohero per-unit economics crossing zero.

---

Moat — AUTO1 Group SE (AG1)

Figures converted from EUR at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Verdict: Narrow moat, strengthening — and unusually physical for a "tech" platform. AUTO1 has a real, mechanical competitive advantage built from four reinforcing assets: a proprietary dataset of actual transaction prices, the largest cross-border used-car logistics network in Europe, a balance sheet large enough to trade metal at continental scale, and a demand-side dealer flywheel with measurable cohort stickiness. The single most important thing to understand is what this moat does: it protects volume, share, and survival — it does not lift the business into high margins. A principal car trader's structural economics are thin by design, so the moat makes the units defensible without changing the fact that the model only works at scale. The strongest real-world proof of durability is not a slide: two better-funded online challengers, Cazoo and Carnext, exited most of AUTO1's markets in 2022 while AUTO1 kept trading profitably through the same downturn [10].

This tab builds on the Business and Industry tabs (which establish the two-engine economic model and the $800bn fragmented arena) and asks the only question that matters here: what protects the returns, how do we know it works, and what would make it fade?

Sources: EU market share 3.1% with a 10% long-term target [15]; ~54,000 unique buying dealers [5]; 6M+ transaction dataset over 14 years [4]; GPU CAGR derived from Merchant GPU $848→$1,147, FY2021–25 [8].

1. The moat scorecard — four pillars, ranked by how much they actually protect

Each pillar is rated on a single discipline: does it produce a cost, data, distribution, or switching edge a well-funded competitor cannot easily copy — and has it survived stress? I deliberately separate the defensible pillars (data, network, scale) from the emerging ones (dealer flywheel, captive finance) and from what is not a moat (the Autohero brand, good execution).

Sources: data exclusivity and balance-sheet/network moat statements [1] [2]; network scale [4]; dealer flywheel and basket growth [6]; captive finance economics [14].

2. The data advantage is the one that is genuinely hard to copy

Management's most credible moat claim is about data exclusivity, and the mechanism is precise rather than hand-wavy. Any AI pricing model is only as good as its data, and "the most important data you can own is pricing data… used car pricing data is private. In order to generate it, you have to start trading. Even classified platforms… do not own the final transaction price data. They only store asking prices and they lack detailed information on the car's condition" [1]. That distinction — settled transaction prices with condition data versus asking prices — is the whole point: a competitor cannot scrape its way to AUTO1's 6M+ proprietary transactions accumulated over 14 years; it must trade for them first, at a loss, against an incumbent that already prices better [4].

This data feeds a concrete economic function, not a brochure. AUTO1's own risk system uses algorithms to "analyse the expected GPU, selling speed, inventories held and market trends" before a car is purchased, declining to buy or diverting hard-to-sell cars to auction — a proprietary underwriting edge in inventory management [12]. Pricing accuracy is exactly what protects GPU on a wafer-thin per-car spread; better data is the difference between a profitable trade and a write-down.

The honest counter: data scale shows diminishing returns. Pricing a Golf in Germany does not require 6 million observations — perhaps a few hundred thousand. The data moat is therefore real at the margin and in long-tail/cross-border models (where thin data hurts rivals most) but is not infinite; a focused competitor in one liquid market could price competently. This is one reason the moat is narrow, not wide.

3. The physical network is the moat management leans on hardest — and it is real

The argument is blunt and, on the evidence, correct: a car weighs one to two tonnes and "cannot be conquered by an AI prompt" — owning the physical pipe (sourcing branches, refurbishment, cross-border logistics) is what lets AUTO1 move metal cheaper than anyone, and "the sheer size of our balance sheet and the efficient management of it through our real-time trade system is another rock-solid element of our competitive moat that simply cannot be replicated by AI software only" [2]. The asset base behind this — 170+ logistics centres, 12 production centres with 248,000 units of refurbishment capacity, 750+ drop-off branches — took more than a decade and hundreds of millions of dollars to build [4].

The cleanest evidence the network works is the steady grind-down in cost-per-car against a rising gross-profit spread — the operating leverage that an integrated owner can capture and a listings site cannot:

Source: derived from Capital Markets Event 2026 — Merchant per-unit gross profit ($848→$1,147) [8], with cost per car summing the marketing-per-unit [18] and operations-per-unit track records [19], FY2021–25.

The widening gap between the blue bar (GPU) and the orange bar (cost per car) is the moat converting to dollars: as volume fills the fixed network, the spread per car drops to profit. Management targets Merchant GPU of $1,238–1,375 long-term against a falling cost base — i.e. the gap keeps widening [8].

4. The demand-side flywheel: dealer stickiness you can measure

The most under-appreciated pillar is a genuine two-sided network effect on the wholesale platform. More buying dealers make AUTO1 the best place for consumers to sell (better prices, faster sale), which deepens supply, which attracts more dealers — the "Merchant flywheel" management diagrams explicitly [7]. Crucially, this is not just an assertion: AUTO1 discloses cohort data showing dealers buy more the longer they stay, and that merchant-finance customers grow their baskets 40–60% — a measurable switching/engagement effect rather than a vague "stickiness" [6].

Source: Capital Markets Event 2026 — unique buying dealers and cohort basket growth [6].

Why this matters for durability: a dealer who has integrated AUTO1 into daily inventory-buying — and who finances purchases through AUTO1 — faces real friction switching to a sub-scale rival with worse selection, worse prices, and no financing. The switching cost is not a contract; it is lost selection, worse pricing, and a severed credit line. That is a softer moat than a regulated licence, but it is exactly the kind that compounds: the captive-finance attach (NIM ~5–7.5%, consumer attach 40%, dealer attach 17% and rising) converts a transactional relationship into a multi-year lending one [14].

5. Does the moat show up in the numbers? Mostly yes — but read it correctly

A moat must appear in returns, margins, share, or retention. Here the evidence is company-specific, not just an attractive industry — the industry lifts everyone equally, but AUTO1 is taking share and expanding margin while the field consolidates.

• Share gains: 3.1% of the European market in 2025, up ~50bps in a single year, against a 10% long-term target — in a market where the top 20 dealers hold under 6%, gaining share is direct evidence the integrated model out-competes the long tail [15].
• Margin expansion in the profitable engine: Merchant adjusted EBITDA reached $281m at a 3.7% segment margin in 2025, up from 3.1% the prior year — the moat is widening, not merely holding [16].
• The caveat that keeps it narrow: group ROCE only reached 6.8% in 2025 — positive, improving, but still below a double-digit cost of capital (per the Business and Financials tabs). A wide moat throws off high returns on capital; AUTO1 does not yet. The moat protects the franchise; it has not yet produced excess returns.

The durability test that matters most is whether the moat held through a downturn. It did. When post-COVID used-car demand normalised and volumes fell in 2022–23, the profit engine stayed profitable every year — the dip is visible but never negative — while two well-capitalised pure-play challengers folded:

Source: Capital Markets Event 2026 — Merchant segment adjusted EBITDA, FY2021–25 [16].

6. What would make the moat fade — and the signals that warn first

Honest moat work weighs the bear case as hard as the bull. AUTO1's own filings name the threats, and the structural ones are real.

Sources: latent entrant threat (Amazon/OEMs) and "highly competitive" framing [11]; AI/classifieds and no-new-entrants view [3]; captive-finance impairment [17]; finance economics [14].

The most important nuance for the bear case: AUTO1's own FY2022 report is candid that the sector is "highly competitive" and that incumbents — independent dealers, classifieds, rental fleets — remain rivals, with Amazon and OEMs as latent entrants holding "significant resources" [10] [11]. The moat is deep enough to defeat under-capitalised start-ups (proven) but untested against a determined balance-sheet equal (a VW Group or Amazon). That gap is the honest ceiling on the rating.

The watchlist that would tell you the moat is fading, in order of read-speed:

First signal — Merchant GPU and segment margin. The clearest moat gauge. If GPU stalls or the 3.7% segment margin reverses while volumes grow, the pricing/cost edge is eroding [16].

Second — dealer cohort retention and basket growth. If older cohorts stop buying more over time, the demand-side flywheel is weakening [6].

Third — cost of credit in the captive book. A rising loss rate (after the $13.9m 2025 lapse) would turn the finance "moat" into a liability and signal the lending build is outrunning underwriting discipline [17].

Fourth — a credible new entrant. Management currently sees none on the sourcing or retail side; the day a well-funded principal trader or OEM enters at scale is the day the durability thesis must be re-underwritten [3].

7. The verdict

The moat thesis and the operating-leverage thesis are, ultimately, the same thesis: owning the physical pipe is what makes the volume defensible, and only scale turns that defensible volume into acceptable returns. The moat is the reason to believe the volume is durable; it is not, on its own, the reason this becomes a high-return business. That distinction is the difference between a narrow rating and a wide one.

---

Financial Shenanigans — AUTO1 Group SE (AG1)

Figures converted from EUR at historical FX rates — see data/company.json.fx_rates for the rate table. Ratios, margins, and multiples are unitless and unchanged.

Forensic Risk Score: 57 / 100 — Elevated. AUTO1's audited income statement crossed into profit in FY2024 and FY2025, yet the business burned more cash than ever in the same two years. That single contradiction — record reported profit alongside a record operating cash outflow — is the spine of this memo. The audited IFRS statements themselves are reasonably faithful: they show the cash burn plainly and classify the asset-backed funding correctly. The forensic risk lives one level up, in how management frames that reality for investors — a "no corporate debt," "self-funding cash generation," "capital-light" narrative that nets roughly $1.9bn of consolidated securitization debt and $400m of annual financing draws out of the headline numbers.

This is a forensic risk assessment, not a fraud allegation. There is no restatement, no auditor resignation, no regulatory action, and no disclosed material weakness. What there is: a severe and widening gap between accounting profit and cash, a presentation layer that systematically softens leverage and cash-generation optics, a used-car revenue base grossed up by an accounting election, a young captive-finance loan book carried with thin reserves, and a founder-anchored governance structure. The flags are linked, not isolated — which is what moves this from "Watch" to "Elevated."

The verdict in numbers

Source: derived from reported financials — net income $91.6m and revenue $9,602.8m [1]; operating cash flow −$544.1m and free cash flow −$574.2m [2]; FCF/net income −6.27x and net debt $1,593.5m [3].

Top two red flags. (1) Profit you cannot bank. FY2025 net income of $91.6m sits against operating cash flow of −$544.1m [1] [2]; the FY2025 accrual ratio is roughly 21% and free cash flow is −6.3x net income. (2) A funding story dressed as a cash-generation story. Investor materials headline "We Finance our Assets Solely Through Non-Recourse Asset Backed Securitizations and Therefore Have Over Half a Billion Free Cash Available" [5] and a non-IFRS "AUTO1 Cash Flow" of +$233m for 2025 [6], even as audited operating cash flow was −$544.1m.

The cleanest offsetting evidence. The audited primary statements do not hide the burn: the consolidated cash-flow statement reports operating cash flow deeply negative and books the securitization draws as financing inflows (+$560.4m in FY2025) rather than smuggling them into operating cash flow [2]. Capex is genuinely tiny — roughly 0.3% of revenue [3] — and technology spend is expensed, not capitalized, so there is no soft-asset cost-parking. Trade receivables are about four days of sales. On the audited record, AUTO1 is more transparent than its slides.

The one data point that would change the grade. Sustained positive IFRS operating cash flow excluding new securitization draws — i.e., the inventory and captive-finance books ceasing to consume cash — would downgrade this toward "Watch." Conversely, a restatement, an auditor emphasis-of-matter or material-weakness finding, or evidence that captive-finance loan-loss reserves are materially understated would push it toward "High."

The central contradiction: profit up, cash down

The most important forensic test for AUTO1 is the one the playbook calls "income statement versus cash-flow statement," and it fails loudly. Across FY2024–FY2025 the company reported cumulative net income of +$113.3m while generating cumulative operating cash flow of −$772.4m [1] [2]. Profit turned positive precisely as the cash gap widened — the opposite of what a maturing, self-funding business looks like.

Source: net income, FY2022–FY2025 [1]; operating cash flow, FY2022–FY2025 [2].

The mechanism is not fraud — it is working capital. AUTO1 buys cars, refurbishes and holds them, and increasingly lends to the consumers and merchants who buy them. Every unit of growth absorbs cash into inventory and loan receivables before it is recovered. Inventory rose 51.8% in FY2025 (to $1,242.8m) against revenue growth of 30.3% [4] [1]. That is the cash sink. The forensic point is narrow but firm: the reported profit is real accounting, but it is not yet cash, and the gap is growing, not closing. A PM should underwrite the cash statement, not the P&L.

How the cash is actually funded: the securitization engine

Here is the test the playbook calls "boosting operating cash flow with unsustainable activities" (CF4), and the related "inflating cash via disposals" (CF3). AUTO1's reported cash position has held up only because financing inflows — draws on asset-backed securitization (ABS) facilities — refill what operations drain. In FY2025, operating cash flow was −$544.1m and financing cash flow was +$560.4m: the two almost exactly offset [2]. This is structural, not one-year: financing inflows have funded the operating burn every year since FY2022.

Source: cash-flow statement, FY2022–FY2025 [2]. Note the FY2022 +$594.5m investing inflow — a money-market-fund disposal that flattered that year's headline cash movement, not an operating event.

To management's credit, the audited statement classifies these draws as financing — the honest place for them. The problem is what happens in the investor deck. There, the same flows are recast. The FY2023 results presentation literally nets ABS funding against asset growth to assert "Over Half a Billion Free Cash Available" [5]. And on the Q4 FY2025 call, management described funding "the growth in inventory through our inventory ABS and also positive trading cash flow," promising "AUTO1 trading cash flows to be positive in 2026 … reinforcing our ability to self-fund our growth" [11]. "Self-fund" is doing heavy lifting for a business whose IFRS operating cash flow was −$544m.

The "AUTO1 Cash Flow" metric: a $777m gap

The clearest single artifact is the Capital Markets Day "AUTO1 has Strong Track Record in Generating Cash" slide. It builds a bespoke "AUTO1 Cash Flow" of +$233m for 2025 — by adding back +$147m of "Inventory ABS Funding" and +$253m of "Captive Finance ABS Funding" against the working-capital outflows [6]. In other words, the metric counts the borrowing as the cash generation. Against the audited IFRS operating cash flow of −$544.1m [2], the company's headline non-IFRS cash number is roughly $777m higher than the statutory figure.

Sources: "AUTO1 Cash Flow" +$233m with $147m inventory-ABS and $253m captive-finance-ABS add-backs [6]; IFRS operating cash flow −$544.1m [2].

This is the textbook key-metric shenanigan: a "cash earnings" label that does not reconcile to the statement of cash flows. It is not illegal and the add-backs are disclosed on the slide — but a reader who anchors on "+$233m generated" rather than "−$544m used" has been steered.

"No corporate debt" versus $1.59bn of net debt

The balance-sheet metric distortion (KM2) is the most quotable. AUTO1 markets a balance sheet with "No Corporate Debt," explicitly excluding its securitization liabilities from a defined "Corporate Net Debt," on the rationale that "FloorPlan financing is generally considered short term payables rather than debt by Ratings Agencies" [7]. On that same slide the inventory ABS liabilities alone are $1,051m, drawn to 87% [7]. On the consolidated balance sheet, total debt was $1,882.2m and net debt $1,593.5m at FY2025, a debt-to-equity ratio of 2.26x — up from 0.92x in FY2022 [3]. The SPVs are consolidated; the debt is on the audited books. "No corporate debt" is a definitional claim, not a balance-sheet fact.

Source: computed net debt and debt-to-equity, FY2022–FY2025 [3]; inventory-ABS liabilities $1,051m at 87% draw [7].

The forensic read: non-recourse is a genuine and material distinction — these facilities are ring-fenced and AUTO1 is right that they differ from corporate bonds. But net debt has grown roughly six-fold in three years while the headline says "none," and leverage has more than doubled. An investor sizing covenant or refinancing risk should use the $1.59bn, not the $0.

Revenue quality: a grossed-up top line

On revenue recognition (EM1/EM2), the headline issue is presentation rather than fabrication. From October 2021, AUTO1 moved warranty and certain remarketing activities from a net (agent) to a gross (principal) basis, recognizing the full transaction value as revenue [9]. This inflates the reported top line without adding economic substance — the same reason gross margin is structurally thin (12.1% in FY2025) [3]. It is defensible under IFRS for a principal, but it means revenue growth overstates economic growth, and gross profit (not revenue) is the metric to track.

The auditor independently flagged revenue recognition as a key audit matter in the FY2021 report — "Die Realisierung der Umsatzerlöse" — alongside inventory valuation [8]. That is appropriate scrutiny, and the audit was unqualified. On the substance: this is a high-volume, largely cash-validated marketplace; trade receivables are about four days of sales, and there is no sign of channel-stuffing or bill-and-hold. The flag is the gross-up, not bogus revenue.

Source: revenue and gross margin, FY2022–FY2025 [1] [3]; net-to-gross revenue election [9].

Reserves, the captive-finance book, and "one-offs"

The hiding-losses test (EM5) is the live one to watch. AUTO1 is scaling consumer and merchant lending, and a young, fast-growing loan book is exactly where under-reserving hides. In the FY2021 accounts the company set inventory valuation and the expected-credit-loss (ECL) approach on financing receivables as focal judgment areas — inventory recoverability was a named key audit matter as used-car prices fell [8]. By 2025 the book was large enough that management recognized $13.9m of merchant-finance credit losses, which it characterized as a one-off learning event [10]. The forensic question is whether a "one-off" credit charge on a portfolio that is still growing is really non-recurring, or an early read on reserve adequacy. This is a medium-materiality yellow flag today; it becomes the thesis if delinquency normalizes higher as the book seasons.

A related, smaller item (EM4): on the Q4 FY2025 call management confirmed that internal refurbishment value-add is capitalized into inventory carrying value — the "$83m" reflected in inventory changes [16]. Capitalizing internal labor into inventory is permissible and common, but it lifts gross profit per unit and carrying value; it is worth monitoring that refurbished inventory clears at those carried values rather than being written down later.

Adjusted metrics and the "record" framing (KM1)

AUTO1 leads its communications with adjusted EBITDA and, from 2024, an "Adjusted Net Income" metric, and brands consecutive years as "record." For 2025 the company points to adjusted EBITDA of roughly $233m [6]; adjusted EBITDA has been a steering KPI since the IPO-year accounts [15]. The gap between adjusted EBITDA and reported net income ($233m vs $91.6m in FY2025) is mostly real D&A, interest and tax — not egregious by software-company standards, and reported net income is now positive, which dampens the usual adjusted-EBITDA concern. The far larger metric-hygiene issue is the cash metric dissected above, not the EBITDA bridge. The debut of a new "Adjusted Net Income" APM in 2024 — exactly when reported results were marginal — is a classic flag, but a low-materiality one here.

Breeding ground: founder-anchored, but not opaque

The governance backdrop amplifies rather than dampens the accounting flags — modestly. AUTO1 runs a two-tier German structure. The Supervisory Board chair is co-founder Hakan Koç, classified non-independent because he is a significant shareholder and former co-CEO; the CEO is co-founder Christian Bertermann [12]. Founder concentration plus a non-independent chair is the structural condition under which optimistic framing goes unchallenged — and the cash-narrative framing above is exactly that. On the dampening side, five of six supervisory directors are independent, the audit & risk committee is chaired by an experienced outsider, and the CFO transitioned in January 2026 [12]. Related-party dealings (founder holdings, the SoftBank vehicle, and a CEO convertible-bond conversion) are disclosed in the related-party note [14]. The auditor has served only since the FY2021 IPO — short tenure, which cuts both ways [13]. Net: a breeding ground that makes the framing risk more credible, without the audit-quality or related-party red flags that would escalate it.

The 13-category scorecard

Every shenanigan family below was tested against the multi-year record. Three categories are red, five yellow, five clean. The reds cluster — CF4, KM1, and KM2 are three faces of one issue: a securitization-funded business presented as a self-funding cash generator with no debt.

Source: severity and evidence per category derived from the cited filings — financials [1] [2] [3]; FY2021 Annual Report revenue & KAM notes [8] [9]; investor-presentation cash & debt framing [5] [6] [7]; captive-finance credit [10].

What to underwrite next

1. IFRS operating cash flow ex-securitization. The single most important number is whether operating cash flow turns positive without new ABS draws. Track operating cash flow against financing cash flow each period [2]; a quarter where operating cash flow is positive on its own merits is the upgrade signal.
2. Captive-finance loan-loss coverage. Monitor ECL/allowance against gross financing receivables and the trajectory of merchant and consumer delinquencies. A second "one-off" credit charge after the $13.9m of 2025 would confirm chronic under-reserving [10].
3. ABS facility draw levels and refinancing. Inventory ABS was already 87% drawn at Q1 2026 [7]; headroom and refinancing terms are the liquidity choke point given the operating burn. Watch facility renewals and any rise in funding cost.
4. Inventory carrying value and refurbishment capitalization. Confirm the $83m of capitalized refurbishment clears through margin rather than reversing as write-downs [16].
5. Metric definitions. Watch whether "AUTO1 Cash Flow," "trading cash flow," and "Corporate Net Debt" definitions stay constant or drift further from IFRS — a moving definition is itself the signal [6].

The decisive read. AUTO1's accounting risk is a valuation-and-position-sizing limiter, not yet a thesis breaker. The audited statements are faithful and unqualified; there is no restatement, no auditor or regulatory action, and the securitization is correctly classified where it counts. But the company is a securitization-funded, working-capital-hungry growth business that markets itself as a debt-free, self-funding cash generator — and that gap is wide, persistent, and structural. An investor should underwrite the −$544m IFRS operating cash flow and $1.59bn of net debt, not the "+$233m AUTO1 Cash Flow" and "no corporate debt," size the position for the funding dependency, and demand a margin of safety that reflects reliance on continued ABS access. If operating cash flow turns genuinely positive ex-ABS and the loan book seasons without further credit surprises, the grade — and the discount — should fall.

---

Figures converted from euro at historical FX rates — see data/company.json.fx_rates for the rate table. Ratios, margins, percentages, share counts, and multiples are unitless and unchanged.

People and Governance — Do Management and the Board Deserve Trust?

Verdict up front: a qualified yes — grade B. AUTO1 is run by its founders, who hold roughly a fifth of the company and are paid almost nothing in cash, so their wealth lives and dies with the share price alongside outside shareholders. The two-tier German structure separates the CEO from the board chair, and five of six supervisory directors are independent [1]. What keeps this from an A is concentration and opacity: co-founder Hakan Koç chairs the very board meant to supervise his co-founder CEO, founder equity has been topped up through capital increases that diluted outsiders, and AUTO1 invokes a German statutory exemption to avoid disclosing individual management-board pay in full [6]. Trust here rests on alignment, not on disclosure or board independence in the formal sense.

The people running the company

AUTO1 is a two-tier German SE: a Management Board (Vorstand) runs the business, and a Supervisory Board (Aufsichtsrat) oversees it. The operating team is founder-led and unusually stable at the top, with one major recent change — a new CFO from outside the company.

Source: Supervisory Board composition, 2026 [1]; FY2021 Annual Report, Related Party Disclosures (Vorstand/Aufsichtsrat) [5]. Holdings as currently reported in market ownership filings (see Ownership).

Two things stand out. First, capability and continuity are genuine — Bertermann has run AUTO1 since founding it in 2012, and the company has scaled from 230 cars in 2012 to 842,000 units sold in 2025 [13]. Second, succession depth is the soft spot. The CFO seat just turned over after Markus Boser's ten-year run, and the new CFO Christian Wallentin — a credible financial-services operator from Hoist Finance — is still bedding in. The CEO role has no obvious internal heir, which is the typical fragility of a founder-driven company.

Ownership and alignment — control sits with the founders and SoftBank

This is the single most important governance fact about AUTO1: it is effectively controlled by a small group of insiders and one large financial sponsor. As far back as FY2021–FY2022, three holders each exceeded 10% of the voting rights — BM Digital GmbH (Bertermann's vehicle), HKVV GmbH (Koç's vehicle), and SVF Midgard / SoftBank Vision Fund [3]. That structure has loosened only modestly since.

Source: structural above-10% holdings (founder vehicles + SoftBank) per FY2022 Annual Report, Takeover-related Disclosures [3]; current percentages as reported in third-party ownership data (not in the filing corpus).

The two founders together still hold roughly 21% of AUTO1 — Bertermann ~12.4% (about $839m at current prices) and Koç ~9.1% (about $611m). SoftBank, which entered in 2018 with ~20% for $521m, has trimmed to ~14.8% but remains the largest single holder. Cadian Capital and Coronation round out a concentrated register, leaving a free float under half the company.

Source: founder stakes per current ownership reporting (not in corpus); CEO cash and variable pay, FY2022 Annual Report, Management Board Remuneration [2].

For an outside shareholder this is the heart of the bull case on people: the founders' incentive is overwhelmingly the share price, not the pay package. Bertermann's roughly $839m equity stake dwarfs his half-million-dollar salary by three orders of magnitude. There is no realistic scenario where management gets rich while the stock languishes. The flip side is control: with founders plus SoftBank near 36% and a founder chairing the supervisory board, minority holders have limited ability to force change.

Compensation — almost nothing in cash, everything in equity

AUTO1's pay philosophy is the opposite of the typical large-cap problem. The cash is small and flat; the leverage is in legacy equity programs.

Source: FY2022 Annual Financial Statements, Total remuneration of the Management Board and Supervisory Board [2].

In FY2022 the entire Management Board — CEO and CFO combined — earned $1,088k, split evenly at $544k each, with zero variable remuneration [2]. The Supervisory Board cost $505k. For a company that did $7.0bn of revenue that year, this is strikingly lean — and it held flat versus FY2021. Across the loss years of 2022–2023 and into the FY2025 turnaround, founder cash pay barely moved.

The equity side is where the real value — and the dilution — sits. The legacy Long-Term Incentive Plan 2020 for the management board comprised 7,500,000 subscription rights at a weighted-average exercise price of $17.85 [8], and a further management-board incentive-share program had 2,064,746 shares outstanding at an $11.10 weighted-average price [9]. A new Remuneration System 2025 and LTIP 2025 were approved to underpin a fresh five-year contract for Bertermann — so equity, not salary, remains the lever.

Pay-versus-performance, judged honestly: because cash pay is fixed and tiny, there is no perverse "pay rose while results fell" dynamic to flag. Management was paid the same modest cash through the deep losses of 2022–2023 as through the record FY2025. The chart below shows the operating backdrop that the founders' equity stakes are actually levered to.

Source: revenue and net income from reported financials, FY2022–FY2025 (company filings, as reported); FY2025 record adjusted EBITDA of $233m confirmed on the Q4 FY2025 earnings call [13].

FY2025 was the inflection: AUTO1 turned its first full-year net profit ($92m) on $9.6bn of revenue and delivered a record $233m of adjusted EBITDA, up 81% [13]. Management's own cash compensation did not spike to capture that turn — a point in their favor.

Board quality and independence — independent on paper, founder-anchored in practice

The supervisory board today numbers six, of whom five are classed as independent; the sole non-independent member is Chairman Hakan Koç, the co-founder [1]. The two-tier structure gives genuine separation of chair and CEO. But the chair being a controlling co-founder means the board's formal independence overstates its practical ability to challenge the founders.

Source: Supervisory Board composition and director backgrounds, 2026 [1].

On expertise the board scores well. Vice-Chair Lars Santelmann ran Volkswagen Financial Services and chairs the Audit & Risk Committee — exactly the automotive-finance depth AUTO1's growing captive-finance book needs. New 2026 addition Jörg Pietzner, head of group accounting at Deutsche Börse, adds genuine financial-reporting and governance heft. Christian Miele (Headline VC) and Claudia Frese (ex-STRATO/IONOS) bring venture and e-commerce perspective [1].

What's notable is how much the board has turned over. The FY2022 supervisory board was a marquee, sponsor-heavy roster — chaired by Dr. Gerhard Cromme (author of the German Corporate Governance Code) and including Gerd Häusler, Vassilia Kennedy and VC Andrin Bachmann (Piton Capital) [10]. By 2026, all of those names are gone and co-founder Koç has moved into the chair. The upgrade in finance/audit expertise is real; the loss of an independent heavyweight chair like Cromme — replaced by the controlling founder — is a governance step backward.

Governance risks and related-party dealings

This is where the words are weighted, because it is where the tension lives.

Three related-party / disclosure points an outside shareholder should weigh:

1. Individual management pay is not fully disclosed. AUTO1 invokes the §286(4) / §314(3) HGB exemption to avoid itemising key-management remuneration in full [6]. In FY2021 total key-management compensation was $3,184k (vs $4,858k in 2020) [5], but the granularity falls short of best-practice say-on-pay regimes.

2. Conflicts are at least handled procedurally. When the supervisory board resolved on the LTIP 2017 — of which Koç was a beneficiary — he was recused from those decisions as a former management-board member and beneficiary [7]. That is the right process; the concern is that the person now chairs the board.

3. Ongoing dilution risk is structural. Total potential dilutive shares from the stacked incentive programs reached ~14.1m as of end-2021 [14], and a new LTIP 2025 extends the pattern. Equity is the founders' reward mechanism, so equity-funded compensation will keep nibbling at the float.

On the credit side, capital discipline is conservative: management runs the balance sheet with roughly $705m of cash and no corporate debt, funding inventory through ABS rather than equity raises, and states the business now self-funds its growth [12]. That removes the most common minority-shareholder fear at a high-growth platform — repeated dilutive cash calls. No supervisory-board loans or advances were granted [2], and the audit relationship with KPMG is long-standing.

For context on shareholder value creation: AUTO1 listed in February 2021 at $43.03 a share, a ~$8.9bn market cap, with Sequoia and Lone Pine as cornerstone investors [11]. The shares now trade around $28 — well below the IPO price — so even the founders, for all their alignment, have not delivered for IPO-era buyers. Alignment cuts both ways: insiders have ridden the same drawdown.

The verdict — Grade B

AUTO1's management deserves a fair degree of trust, earned the hard way: through skin in the game rather than through disclosure or formal independence. The founders own ~21% of the company, take almost no cash, run a debt-free, self-funding balance sheet, and have steered the business to its first profit and record EBITDA in FY2025. The board carries real automotive-finance and accounting expertise.

The reasons it is a B and not an A are concrete and concentrated: a controlling co-founder chairs the supervisory board he is meant to be supervised by; individual executive pay hides behind a statutory exemption; and founder equity has repeatedly been topped up through dilutive, related-party share issuances. None of these is a fraud flag — all are disclosed — but together they tilt power decisively toward insiders.

---

Figures converted from euros (EUR) at historical FX rates — see data/company.json.fx_rates for the rate table. Ratios, margins, multiples, unit counts, and percentages are unitless and unchanged.

History — How the Story Changed: From $43 Hype to a Quiet Profitability Grind

AUTO1 came public in February 2021 selling a hyper-growth dream — Europe's leading used-car platform, with the consumer-retail brand Autohero as the crown jewel that would "become the leading retailer for used cars in Europe" [1]. Within eighteen months that story had broken: cash burn peaked, the share price collapsed from a $43.04 IPO price [3], and in Q2 2022 management quietly inverted the entire narrative — from growth-at-all-costs to profitability-first [5]. What followed is the most important fact about this management team: they did almost exactly what they then said they would do, quarter after quarter, while saying remarkably little about the original promises they had stopped being able to keep. Credibility on near-term execution has improved dramatically since 2022; credibility on the grand IPO promises is still unproven five years on. This page traces that split.

Who built this, and when the current chapter began

This is a founder-run company, not a turnaround under hired managers. Christian Bertermann and Hakan Koç founded AUTO1 in 2012 [2]; Bertermann has been CEO throughout and became sole CEO in January 2021 when Koç stepped up to chair the Supervisory Board. So there is no "inherited" business to disentangle — this team built every part of it, and at the IPO the consolidated group was deeply loss-making, not a high-quality asset handed over. The one genuine pre-existing jewel they had built was the Merchant (wholesale) business, which was already near cash-generative; the rest — Autohero retail and the lending arm — was, and partly still is, a work in progress.

The present strategic chapter began in Q2 2022, when management pivoted from chasing units to chasing profit — the single hinge on which this whole history turns.

Source: leadership and chapter anchors per IPO announcement [2] and the Q3 FY2023 call confirming the Q2 2022 profitability pivot [5]; first-profit year derived from reported financials.

The narrative drifted — here is what management stopped saying, and started saying

The fastest way to see how the story bent is to track which themes management emphasized each year. "Growth at all costs" — the unspoken 2021 default — was scrubbed from the script by 2023 and replaced by "breakeven" and "unit economics." Once profit was proven, the long-dormant 5–9% margin and 10% market share ambitions reappeared, and a fresh "AI / data moat" narrative was layered on top in 2025–26 to defend the equity story against large-language-model disruption fears.

Source: author's coding of management emphasis across the earnings-call record, FY2023–Q1 FY2026 — e.g. the profitability pivot [5], the 5–9% margin target's reappearance [10], and the AI-moat framing [18].

Chapter 1 — The IPO promise and the break (2021–2022)

The January 2021 IPO pitch was unambiguous: a roughly $680 billion European used-car market, an offline-to-online shift, and a plan to deploy ~$850 million of the proceeds to "accelerate the growth of its business, especially of Autohero" and make it "the leading retailer for used cars in Europe" [1]. The shares priced at $43.04 [3], valuing the company near $8.9 billion.

The economics did not cooperate. FY2022 brought the deepest losses in the company's history — a net loss of $262m and roughly $469m of negative free cash flow as Autohero was scaled into a rising-rate, falling-used-car-price environment. The growth-at-all-costs model was structurally unfinanceable at that valuation, and the equity de-rated hard. This is the "break" — and it set up the pivot.

Chapter 2 — The pivot: profit over growth (Q2 2022 → 2023)

In the spring of 2022 management did something founders rarely do voluntarily: they chose to grow slower. They deliberately shrank the top line — group revenue fell from $6.97bn (FY2022) to $6.04bn (FY2023) — to force gross-profit-per-unit (GPU) and cost discipline. On the Q1 2023 call, CFO Markus Boser reported an adjusted-EBITDA loss of $28m and pledged group breakeven "by Q4 this year" [4].

They beat it. By Q3 2023 the group had "reached EBITDA breakeven ahead of plan," posting positive adjusted EBITDA of $0.6m — one quarter early — and Bertermann dated the turn precisely:

"since we shifted our focus towards profitability in Q2 of last year"

— pinning the start of the current chapter to Q2 2022 [5]. Delivering a self-imposed profitability target early, right after a credibility-destroying IPO, is the first and most important deposit in this management's trust account.

Source: revenue and net income per reported financials, FY2022–FY2025 [17]; FY2021 revenue from segment disclosure (company filings, as reported).

Chapter 3 — Earning the right to talk about the future again (2024)

2024 was the payoff year. It opened with a then-record $18m of adjusted EBITDA in Q1 [7] and built to "a landmark quarter," 177,000 units and $35m adjusted EBITDA in Q3 [8] — and the year the company posted its first annual net profit (+$22m). The most revealing moment came when an analyst noticed the IPO-era 5–9% long-term margin target had reappeared in the appendix after years of absence. Boser's answer is the cleanest articulation of this team's self-awareness:

"from a capital markets perspective, we're blunt, we were a show-me story… We kind of feel now that we are showing you"

[9]. The long-term 5–9% adjusted-EBITDA-margin ambition was formally back on the slides [10] — note that it had effectively gone unmentioned through 2022–23, the classic narrative-drift tell: the goal you stop repeating until you can back it up.

Through 2023–24 management also reframed the Merchant gross-profit-per-unit guidance upward and beat it every time. It started at "$702 to $736 at IPO," was raised to $831–935 by Q1 2024 [6], and again to $857–961 by Q3 2024 [11]. This is the boring promise they kept relentlessly.

Chapter 4 — Re-acceleration, records — and the questions a record year raises (2025–26)

With unit economics fixed, the team flipped back to growth — profitably this time. FY2025 was a genuine record on the metrics management chooses to highlight: 842,000 units, $233m adjusted EBITDA (up 81%), a 2.4% margin — the best in the company's 14-year history [17], with market share reaching a record 3.1% [21]. Almost every quarter through the year reset the record — $68m of adjusted EBITDA already in Q1 2025 [14]. Bertermann reframed AUTO1 as an "AI-enabled Amazon for the used car market," leaning hard into proprietary pricing data as the moat [18].

Yet the market's reaction to that "fantastic year" was a sharp sell-off on results day (25 Feb 2026), with the stock falling roughly 20% intraday — because the 2026 guidance implied margin progress stalling: $287–315m adjusted EBITDA on a renewed spending push, and $69m of Q1 2026 EBITDA that was only $2m above the prior year [22], [19]. Five years after a $43 IPO, the shares sit near $27.

The tell hiding under the records: a record profit alongside record cash burn

The single most important thing a skeptic should hold onto is this: FY2025 produced the company's best-ever reported profit and its worst-ever free cash flow (roughly −$575m). Management resolves the apparent contradiction by separating "trading cash flow" (positive, self-funding) from investment in the on-balance-sheet captive finance book — consumer and merchant loans — which grew almost 50% in 2025 and is funded through asset-backed securitizations [19]. That framing is reasonable, but a reader should not let "self-funding growth" obscure that AUTO1 is increasingly a balance-sheet lender whose reported earnings outrun its GAAP cash generation.

Source: net income and free cash flow per reported financials, FY2022–FY2025 (company filings, as reported); FY2025 cash dynamics discussed on the Q4/FY2025 call [19].

The deeper tell: the whole group profit is Merchant; Autohero — the IPO crown jewel — still loses money

At the June 2026 Capital Markets event (its first ever, five years after listing) AUTO1 finally disclosed segment-level P&Ls. They tell the real story. Merchant adjusted EBITDA climbed from $70.9m (FY2021) to $281m (FY2025) [29]. Retail / Autohero — the brand the IPO was built around — was still adjusted-EBITDA negative in FY2025 at −$48.9m (−2.4% margin) [26], six years after launch. The losses have narrowed impressively (from −$192m in FY2021), but the crown jewel has yet to earn its keep.

Source: Capital Markets Event 17 Jun 2026 — Merchant segment P&L [29] and Retail segment P&L [26].

Management's honesty about Autohero is itself a credibility marker worth weighing. They have not claimed retail profitability they do not have. In Q1 2025 they flagged reaching segment profitability "in at least one of the months of Q1" — and immediately chose to spend it back on growth [13]. By Q4 2025 the careful wording was:

"Retail unit economics on a segment allocated base for the full year of 2025 are positive before marketing"

[20] — "before marketing" being the load-bearing qualifier, and Boser confirming the segment remains loss-making after it [16].

Retail GPU: the one long-term promise that has visibly tracked toward its target

The $3,440 long-term Autohero gross-profit-per-unit target, first set years ago, is the rare grand promise that has marched steadily toward delivery — from $1,491 in Q1 2023 to $3,130 by Q3 2025 — and management still reaffirms it [6].

Source: Autohero retail GPU as reported on quarterly calls, Q1 FY2023 [5] through Q1 FY2026 [23]; €3,000 (≈$3,440) target reaffirmed on the Q1 FY2024 call [6].

The credibility verdict — evidenced

Split the promises into two buckets and the picture is sharp.

Near-term, self-set operating guidance: kept or beaten, repeatedly. Group EBITDA breakeven delivered a quarter early [5]; first net profit delivered on schedule in 2024; Merchant GPU guidance raised and beaten through the cycle [6]; full-year EBITDA guidance raised multiple times in both 2024 and 2025 and beaten each year [19]. And misses are handled honestly — the "show-me story" admission [9], the disclosure that a 2025 Merchant-receivables impairment ($13.9m) came from a self-inflicted underwriting lapse in one market [24], and the "before marketing" precision on retail all point to a team that does not spin.

The grand IPO promises: still open, five years on. The 5–9% margin (2.4% today) [12], the 10% market-share goal (3.1% today) [28], Autohero as a profit engine (still loss-making) [26], and IPO-era expectations of self-funding cash generation (GAAP FCF deeply negative) remain to be earned. The June 2026 segment targets — Merchant adjusted EBITDA to $550–826m and Retail to swing positive [25], [27] — are the next test.

Source: promise-vs-delivery track record compiled from the earnings-call and Capital Markets record — e.g. early breakeven [5], GPU guidance walk [6], long-term targets [12], and segment economics [26].

What the story is now — believe vs. discount

Believe: the Merchant business is a real, scaled, cash-generative European wholesale platform that compounds GPU and units, and a founder-led team that has earned the right to be taken at its word on near-term targets. The CFO handover from Markus Boser (after a decade) to Christian Wallentin, a career financier, in January 2026 [15] is sensible given how central the captive-finance book now is.

Discount: the equity story still rests on two unproven legs — Autohero turning from a six-year cash sink into the promised profit engine [27], and the group reaching a 5–9% margin from 2.4% — and the growing on-balance-sheet lending arm means reported profit and real cash generation are diverging, not converging.

Is the story simpler and more durable than it was at IPO? Yes — it is no longer "fund a moonshot." Is it de-risked enough to take the long-term targets on faith? No. Credibility on execution is clearly improving; credibility on the decade-defining promises is still pending its verdict. The 2026–27 margin trajectory, and whether Autohero finally crosses into segment profit while still growing 30%+, will decide which of the two AUTO1 stories — the proven Merchant compounder or the still-unproven retail-and-lending dream — the market ends up paying for.

---

Financials — AUTO1 Group SE (AG1)

Figures converted from EUR at historical FX rates — see data/company.json.fx_rates for the rate table. Ratios, margins, and multiples are unitless and unchanged.

AUTO1 is a $9.6bn-revenue digital used-car platform that has just crossed from a decade of losses into profit. The number that matters is not revenue and not net margin — it is gross profit per unit (GPU) and adjusted EBITDA, because AUTO1 books the entire sale price of every car it trades as revenue, so its income statement looks like a razor-thin retailer when the economics are those of a scaling marketplace. In FY2025 the model inflected: 842,271 units sold (+22%), gross profit of $1,164.0m (+37%), and record adjusted EBITDA of $232.1m (+81%), a 2.4% adjusted-EBITDA margin [1]. The balance sheet carries no corporate debt, and the deeply negative reported free cash flow is an accounting artifact of growth funded by non-recourse inventory securitizations [4].

Source: FY2025 Results Trading Update, Group KPIs [1]; net income per reported financials.

---

1. How to read this company's statements

The gross-revenue trap. AUTO1 takes ownership of most cars it sells — both in the Merchant wholesale business (buy from consumers via wirkaufendeinauto.de, sell to dealers via AUTO1.com) and in Retail (refurbish and sell to consumers via Autohero). Because it is the principal in the transaction, IFRS requires it to recognise the full vehicle price as revenue. That is why FY2025 revenue was $9.6bn [14] on a gross margin of only 12.1% and a net margin of under 1%. Reading AUTO1 like a normal retailer — fixating on the 1% net margin — misses the business entirely.

Read these three lines instead:

Gross profit ($1,164.0m FY2025) — what AUTO1 actually keeps after the cost of the car; this is the true "revenue" of the platform.

GPU — gross profit per unit ($1,377 FY2025) — the per-car economics; the single cleanest gauge of pricing power and product attach.

Adjusted EBITDA ($232.1m FY2025) — gross profit minus operating costs before share-based and one-off items ("special defined items"); the profit the model throws off at scale [1].

Defined once, used throughout: GPU = gross profit ÷ units sold. Adjusted EBITDA strips out depreciation, share-based compensation, and restructuring. Trading cash flow is management's term for operating cash flow excluding the inventory and finance-receivable swings that are funded by dedicated asset-backed facilities — the figure that tells you whether the model self-funds (see §4).

---

2. The profitability inflection

For most of its public life AUTO1 burned cash to buy growth. FY2022 carried a $223m operating loss; FY2023 still lost $113m at the operating line. Then operating leverage arrived: gross profit grew far faster than operating costs, and the company swung to a $57m operating profit in FY2024 and $166m in FY2025, with the first-ever full-year net profit of $92m and diluted EPS of $0.41 [1].

Source: reported income statement, FY2022–FY2025; FY2025 figures corroborated in the FY2025 results deck [1].

The deeper signal sits in adjusted EBITDA, which management reports as the cleanest profit measure. It moved from the "best result since IPO" of negative $49m in FY2023 [2] to +$113m in FY2024 and +$232.1m in FY2025 — a $281m swing in two years on essentially flat-to-up unit pricing.

Sources: adjusted EBITDA FY2023 (negative $49m) per press digest [2], FY2024–FY2025 per FY2025 results deck [1]; group units sold reconciled to segment disclosures in the 2026 Capital Markets Event [10].

Note the shape: units dipped in FY2023 (used-car demand normalised post-COVID) yet gross profit held and adjusted EBITDA improved — proof that the inflection was driven by unit economics and cost discipline, not just volume. GPU climbed from roughly $850 in FY2021–22 to $1,377 in FY2025 [1], while operating expense per unit grew far more slowly. That is the definition of operating leverage, and it is the heart of the bull case.

---

3. Segment economics — the engine and the swing factor

AUTO1 reports two segments, and they could not be more different in profitability. Understanding the split is the whole game.

Merchant (B2B wholesale) — the profit engine. This is AUTO1.com, Europe's largest digital wholesale platform for used cars, plus the consumer-sourcing brands. It is structurally profitable and accelerating: segment adjusted EBITDA rose from $37m in FY2022 to $164m in FY2024 and $281m in FY2025, a 3.7% segment margin [10]. Merchant alone earns more than the whole group — because Retail still loses money.

Retail (Autohero) — the swing factor. Autohero buys, refurbishes, and sells cars direct to consumers. It is the loss center, but the loss is shrinking fast on a per-unit basis: adjusted EBITDA per retail unit improved from negative $4,644 in FY2021 to negative $482 in FY2025, with a long-term target of positive $1,704 to $2,832 per unit [11]. Retail GPU has compounded from $410 to $3,100 per unit over the same span [11]. If Autohero crosses into per-unit profit, group adjusted EBITDA re-rates sharply; if it stalls, the group is "just" the Merchant engine. This is the single most important number to track.

Source: 2026 Capital Markets Event, Merchant segment financials [10].

Source: 2026 Capital Markets Event, Retail (Autohero) per-unit economics; long-term target is positive $1,704–$2,832 adjusted EBITDA per unit [11].

A third profit lever sits inside both segments: captive finance. AUTO1 increasingly originates loans to consumers and dealers, lifting GPU. Management targets a 50–60% financing attach rate (40% today), a 5–7% net interest margin, and cost of credit held at 1–2% [12]. This is high-quality, recurring gross profit — but it also means AUTO1 is becoming part-lender, which is why credit losses (cost of credit) belong on every watch list.

---

4. Earnings quality and the cash-flow illusion

This is where most screens get AUTO1 wrong. The reported numbers show operating cash flow of negative $544m and free cash flow of negative $574m in FY2025 — against $92m of net income. Taken at face value, that looks like catastrophic earnings quality. It is not.

Source: reported cash-flow statement, FY2022–FY2025 (as reported).

Why the gap is benign. AUTO1 deliberately grew inventory from $724m to $1,243m during FY2025 to support 22% unit growth [5]. That inventory build is a cash use in the operating line — but it is funded almost entirely by a dedicated inventory asset-backed securitization (ABS) facility, and the cash drawn from that facility lands in the financing line. The standard cash-flow presentation therefore splits one self-funding transaction across two statements, manufacturing a scary operating outflow next to an offsetting financing inflow (financing cash flow was +$560m in FY2025).

Management strips this out with "trading cash flow" — operating cash flow excluding the ABS-funded inventory and finance-receivable swings. On that basis the model was cash-generative in 2025 and is guided to stay positive in 2026: "we did grow with positive cash flow on the trading cash flow… we continue to have free trading operating cash flow in '26… we're self-funding the growth" [6]. Capex is genuinely trivial — about 25 basis points of revenue ($31m on $9.6bn) [4] — so this is an asset-light platform, not a capital sink.

---

5. Balance sheet — no corporate debt, financed by non-recourse ABS

On a screen, AUTO1 shows $1.9bn of total debt and $1.6bn of net debt — leverage that would alarm for a thin-margin retailer. The reality is more reassuring, and it hinges on what kind of debt this is.

Management is emphatic: "no corporate debt" [4]. The debt on the consolidated balance sheet is the inventory and captive-finance ABS — non-recourse funding raised through special-purpose vehicles whose lenders have a claim only on the assets inside the SPV, not on AUTO1 itself [13]. The funded assets (inventory $1,243m, captive-finance receivables $1,001m) and the funding move together; if growth stops, the inventory liquidates and the facilities repay.

The corporate balance sheet is conservatively capitalised: about $710m of total cash at year-end 2025 and $850m of equity by Q1 2026, an equity ratio near 24%, with inventory financed at roughly 80% LTV through non-recourse ABS [5] [8]. Cash actually rose $55m in Q1 2026 to $748m even as the business grew [9]. And the funding runway was just extended: in 2026 AUTO1 raised inventory-financing capacity 45% to $1.8bn with a thirteen-lender syndicate and a revolving period out to November 2027 [7].

Source: reported balance sheet, FY2022–FY2025; "cash" here is the narrow cash-and-equivalents line — management's broader "total cash" was about $710m at YE2025 [5].

One genuine scar: years of losses left an accumulated deficit of roughly negative $1.6bn in retained earnings. Equity of $831m exists only because of the IPO and subsequent capital raises. AUTO1 has earned back none of its historical burn yet — FY2025 was year one of profit. The balance sheet is sturdy today, but the company has no long track record of compounding its own capital.

---

6. The year-wise statements

Sources: reported income statement, balance sheet and cash-flow statement (FY2022–FY2025); FY2021 revenue and group units reconciled to segment disclosures [10]; FY2025 KPIs per results deck [1]. "Total debt" is overwhelmingly non-recourse ABS, not corporate borrowing; "cash" is the narrow reported line (management total cash ≈ $710m at YE2025).

Source: derived from reported financials, FY2022–FY2025.

Both margins rise every year — but remember they are calculated on gross revenue, so the level (12% gross, under 2% operating) is structurally low by design. The trend and the absolute dollars of gross profit are what matter.

---

7. Returns on capital and capital allocation

Returns inflected with profit. Return on equity swung from negative 36% in FY2022 to negative 20% in FY2023 to +3.4% in FY2024 and +11.0% in FY2025; return on capital employed reached 6.8% [1]. For a business in year one of profitability with a Retail segment still in investment mode, an 11% ROE is a respectable start — though it is flattered by the thin equity base left after years of losses.

Source: derived from reported financials, FY2022–FY2025.

Capital allocation is simple and, for now, correct: reinvest everything. AUTO1 pays no dividend and runs no buyback — every dollar of cash and every ABS line goes into funding inventory growth, scaling Autohero, and building the captive-finance book. With Merchant compounding and Autohero approaching per-unit break-even, plowing capital back is the value-maximising choice. Share count is stable (~219m, up only modestly from dilution), so this is reinvestment, not empire-building funded by issuance [1]. The capital-allocation question shifts in 2–3 years: once Retail turns profitable and trading cash flow scales, will management return cash or keep reinvesting? There is no policy yet — appropriate for the stage, but a future governance watch item.

---

8. Valuation — priced for a successful Autohero ramp

At about $28.0 per share (mid-June 2026) AUTO1 is capitalised near $6.1bn. Because the ABS debt is non-recourse and offsets funded assets, the cleaner enterprise value is roughly market cap less the ~$0.7bn corporate net cash — call it ~$5.4bn.

Source: share price per market data (19 Jun 2026); multiples derived from reported financials and FY2026 guidance [3].

On consensus, AUTO1 trades at roughly 39× FY2026 EPS ($0.72) and 26× FY2027 EPS ($1.09), or about 18× the midpoint of FY2026 adjusted-EBITDA guidance ($287–315m) on a corporate-EV basis [3]. Those are growth multiples — justified only if Autohero turns and adjusted EBITDA compounds 25–30% for several years. Nothing here is "cheap"; the question is whether the multiple is supported by the trajectory.

Versus history and the trajectory: AUTO1 has re-rated hard from its 2022 lows as profitability arrived, and analysts are constructive (mean target ~$36 versus ~$28 price; 11 of 14 rate it buy). But the market is unforgiving on the swing factor: when the FY2026 outlook implied gross profit per unit below the Q4 2025 level — the cost of pushing faster Autohero growth — the shares fell about 7% in a day [2]. That reaction tells you exactly where the valuation's sensitivity lives: near-term GPU and the credibility of the Autohero margin path.

Versus peers. The peer screen mixes two different revenue models, and that distinction governs the comparison.

Source: latest reported fiscal-year financials per company filings; market caps as captured in the run dataset. EUR figures converted to USD; Auto Trader shown in GBP (no GBP/USD rate in the run dataset).

The honest read: AUTO1's 12% gross margin and under-2% operating margin look like the gross-revenue retailers (Carvana, CarMax, Aramis), not the high-margin marketplaces (Auto Trader at 62% operating margin, OPENLANE) — because, like them, it books the whole car as revenue. Within that group AUTO1's 30% revenue growth is second only to Carvana, and its 11% ROE sits mid-pack and rising. Carvana is the closest trajectory analog (the operating-leverage-from-scale story that re-rated violently as EBITDA arrived); Aramis is the closest European business-model peer but far smaller and slower-growing. Auto Trader and OPENLANE are fee-revenue marketplaces — useful for AUTO1.com's economics in isolation, but not comparable at the group revenue line. AUTO1 deserves a growth premium to Aramis and CarMax; whether it deserves to trade toward Carvana's multiple depends entirely on Autohero.

---

9. The bottom line

The financials confirm a genuine turnaround: record gross profit, record adjusted EBITDA, the first net profit, an asset-light cost base (capex 0.25% of revenue), and a conservatively funded balance sheet with no corporate debt. They contradict the lazy bear case built on the screen — the "catastrophic" negative free cash flow is an artifact of ABS-funded inventory growth, and management's self-funding trading cash flow is positive [6]. What they cannot yet confirm is that Retail will pay off: Autohero still loses money per unit, the equity already prices its success, and the market punishes any GPU wobble.

The first financial metric to watch is Autohero (Retail) adjusted EBITDA per unit. It improved from negative $4,644 to negative $482 over four years, with a long-term target of positive $1,704–$2,832 [11]. The quarter Retail crosses zero is the quarter the whole equity story is validated; a stall there — especially if cost of credit in the captive book rises — is the fastest way to break the thesis. Watch it alongside group adjusted EBITDA against the $287–315m FY2026 guide [3].

---

Figures converted from euros at historical FX rates — see data/company.json.fx_rates for the rate table. Ratios, margins, and multiples are unitless and unchanged.

Bottom line up top

The filings tell a clean "record everything" story; the web tells you what kind of record it is, and what the market is still arguing about. AUTO1's profit engine is the wholesale Merchant business (≈$281m adjusted EBITDA in 2025); its headline growth engine, the Autohero consumer-retail segment, still lost money at the EBITDA line (≈−$49m in 2025) — and management has explicitly chosen 2026 as a "hold gross-profit-per-unit flat, buy volume" year, so the $57–86m of EBITDA growth comes from units, not margin. That single fact already cracked the stock once (−7% on the 26 Feb / Mar 2026 margin guide) before the June Capital Markets Event rebuilt the bull case. The second thing the web reveals that the income statement hides: AUTO1 is increasingly a financing business — net debt nearly doubled to ~$1.29bn in 2025, free cash flow per share is negative, management declines to guide FCF, and the new CFO comes from a consumer-credit firm. The accounting is otherwise uncontested — no short-seller report, no probe, no auditor flag surfaced anywhere in the public record.

Sell-side is crowded long (consensus mean target ~$35.4 vs ~$28.0 spot, mostly Buys), so the upside here is consensus, not contrarian — the edge is entirely in whether Autohero's margin inflects or keeps the group de-rated.

The numbers the web frames

Sources: price and multiples per Yahoo Finance (AG1.F key statistics, 19 Jun 2026) and MarketScreener (AUTO1 finances, 19 Jun 2026); consensus target per company estimate feed (mean $35.4, median $37.7, range $23.4–$43.6). Net debt per MarketScreener consensus; reconciled to "no corporate debt + ABS-funded inventory" in the FY2025 call [2].

Ranked findings

1. The 2026 "growth-over-margin" pivot is the whole debate — and it already moved the stock both ways

AUTO1's FY2026 guidance (940k–1.0m units, gross profit $1.26–1.38bn, adjusted EBITDA $287–315m) rests on GPU held broadly flat versus 2025, with the EBITDA step-up driven by volume rather than margin [2]. The market read that as a margin disappointment: shares slid ~7% when the 2026 outlook landed because faster (loss-making) Autohero growth implied gross-profit-per-unit below Q4 2025 (Investing.com, 26 Mar 2026). At the 17 Jun 2026 Capital Markets Event, management reset the narrative with long-term targets (Merchant +10–15%/yr, Retail +20–40%/yr) and the stock rallied (Investing.com, 17 Jun 2026).

So-what: this is the swing factor for the entire equity. Bulls own the volume story (1m-unit milestone in 2026); bears own the margin pause. Priced in? Partly — the stock fell to a ~$23.5 / ~$3.9bn-market-cap trough around the March guide (forward P/E ~23x at that point) and has since round-tripped back to ~$28.0. The unresolved question the market has not settled is whether Retail margin inflects in 2027; that, not 2026 units, is what re-rates or de-rates the multiple. Neutral / red flag.

2. The growth engine still loses money — Autohero was ≈−$49m EBITDA in 2025

The segment disclosures aired at the CMD make the internal cross-subsidy explicit: Merchant generated 2025 revenue ~$7.5bn, gross profit ~$850m and adjusted EBITDA ~$281m (3.7% margin), while Retail/Autohero generated ~$2.07bn revenue, ~$315m gross profit and adjusted EBITDA of ~−$49m (Quartr — AUTO1 CMD 2026 summary, 17 Jun 2026). In other words, all of the group's $232m adjusted EBITDA — and then some — comes from wholesale; the high-growth, higher-GPU retail brand is still a cash consumer being scaled into profitability (long-term target $940+ EBITDA/unit on 300k units).

So-what: the bull case is a sum-of-the-parts where a profitable Merchant compounder funds an Autohero option that is years from its target economics. Sizing should respect that the segment growing 20–40% is the one losing money. Priced in? This is broadly understood by the covering analysts but is easy to miss from the consolidated "record EBITDA" headline. Neutral.

Source: AUTO1 2026 Capital Markets Event segment disclosure (17 Jun 2026), as reported via Quartr; group figures cross-checked to the FY2025 earnings call [5].

3. AUTO1 is becoming a financing business — net debt nearly doubled, FCF is negative, and management won't guide it

Consensus balance-sheet data show net debt rising from ~$584m (2024) to ~$1,285m (2025), heading toward ~$1.43bn in 2026 (MarketScreener, finances, 19 Jun 2026). Management's framing is different but reconcilable: on the FY2025 call it stressed "no corporate debt," ~$705m cash, and ~$1.2bn of inventory funded through an inventory-ABS programme at 83% loan-to-value, plus a growing captive-finance receivables book funded by further ABS [2]. The "net debt" the screens show is that ABS-funded inventory and loan book. Critically, when pressed for free-cash-flow guidance for 2026, management declined, pointing only to positive "trading/operating cash flow" and self-funding via ABS [4]. Third-party data put free cash flow at roughly −$2.5 per share (Meyka, 2026). The ABS machine keeps expanding — a second public consumer-loan securitisation, FinanceHero-2 ($278m), priced in Sep 2025 (GlobalCapital, 15 Sep 2025).

So-what: "record adjusted EBITDA" and "negative free cash flow with rising leverage" are both true at once — the growth is working-capital- and credit-intensive, and a chunk of profitability now depends on captive-finance income and continuous access to the ABS market. A credit-market shock or rising delinquencies hit this model where the income statement doesn't show it. Priced in? Under-appreciated — the bull notes lead with EBITDA, not the cash conversion gap. Red flag.

4. Reported Autohero GPU is flattered by ~$83m of capitalised internal refurbishment — an earnings-quality nuance worth watching

On the FY2025 call, a UBS analyst pressed that reported Autohero GPU is not gross profit ÷ cars sold because internal refurbishment costs are capitalised into inventory — a swing worth ~$83m in Q4 2025 alone — and management confirmed the mechanic [3]. Reported Retail GPU of $3,061 (FY2025, target $3,525) therefore overstates the cash margin to some degree [6].

So-what: this is the one place the otherwise-clean accounting deserves scrutiny — capitalising in-house refurbishment shifts cost off the current P&L and inflates reported per-unit gross profit, exactly in the segment whose margins the bull case depends on. Not a red flag on its own (it is disclosed and analyst-tested), but the gap between reported GPU and cash margin is the metric to track. Neutral / watch.

5. Valuation is rich and consensus is crowded long — the contrarian reads are bearish

At ~$28.0 the stock trades at EV/EBITDA ~38.8x, trailing P/E ~76x, forward P/E ~39x, P/B ~7.6x (Yahoo Finance, AG1.F key statistics, 19 Jun 2026) — an independent screen flags it as a "classic high-risk growth play… more expensive than 92% of its industry" with riskier-than-average financing (Obermatt). Yet the sell-side is overwhelmingly positive: in the days around the June CMD, UBS reiterated Buy and raised its target to $38.64, with JP Morgan, Deutsche Bank and Jefferies all on Buy; Berenberg is the lone Hold/Neutral (MarketScreener recommendations, 18 Jun 2026). Coverage was initiated Buy by Jefferies ($40) in Dec 2025 (Investing.com, 9 Dec 2025), after BNP Paribas Exane (→Outperform, $31, Mar 2025) and Goldman (→Buy, Jan 2025) upgrades. The dissent is quantitative: a multiples-based model pegs fair value at $21.1 — ~25% below the price (AlphaSpread), and near-term EPS estimates have been revised down (current-year EPS trimmed to $0.73 from $0.76 ninety days ago).

So-what: mean target ~$35.4 implies ~27% upside, but that view is consensus and the multiple already embeds the growth. The asymmetry is unattractive if Autohero margins stay suppressed — there is little contrarian cushion. Priced in? The bull case is the consensus; the variant view is the bearish valuation models, which are not in the price. Neutral / red flag on valuation.

Sources: MarketScreener analyst recommendations (18 Jun 2026); Investing.com analyst-ratings coverage (Jefferies 9 Dec 2025, BNP Paribas Exane 10 Mar 2025, Goldman 13 Jan 2025).

6. Management and governance: a financing-savvy CFO swap, board churn, founders still in control

The most telling people signal is the CFO transition: Markus Boser stepped down at end-2025 after ten years and was succeeded on 1 Jan 2026 by Christian Wallentin, who joined from consumer-credit firm Hoist Finance (AUTO1 press release, 1 Oct 2025) — a credit/banking background that fits a business pivoting toward captive finance and ABS. Boser sold shares worth ~$0.66m in May 2025 before departing (InsiderScreener); there has been no insider buying in the past six months, and institutions were modest net sellers last quarter (Yahoo Finance insider transactions). The supervisory board has turned over (chairman Gerhard Cromme and Gerd Häusler left in 2024; Jörg Pietzner added June 2026), while co-founders Christian Bertermann (CEO) and Hakan Koç (supervisory-board chair) remain in control. ISS scores governance an overall 6/10, but with Compensation flagged high-risk (8/10) against low shareholder-rights risk (1/10) (Yahoo Finance profile, 4 Jun 2026).

So-what: founder alignment and clean shareholder rights are positives; the comp-governance flag, the absence of insider buying, and a departing CFO trimming his stake are mild negatives, not thesis-breakers. The Wallentin hire is the substantive tell — it signals the financing arm is becoming central, not incidental. Neutral.

7. Forensic all-clear — the public record does not contest the accounting

Targeted searches for a short-seller report, fraud or accounting allegation, regulatory probe, auditor resignation, material weakness, restatement, or litigation returned nothing AUTO1-specific. The most recent filings are routine (a major-shareholding/voting-rights notification dated 16 Jun 2026; AGM results in June 2026), not red flags.

So-what: "the news is that there is no news" here is genuinely useful — apart from the capitalised-refurbishment nuance in Finding 4, the filing-based thesis is uncontested by the public record. The real risks are operational and financial (Autohero margin, cash conversion, ABS dependence), not allegations of misconduct. Positive.

8. Competitive position: structural digital leadership intact, but absolute share is still tiny

External research corroborates the filings rather than contradicting them: AUTO1 holds a ~3.1% share of a fragmented ~$800bn European used-car market (top-20 traditional retailers ≈6% of transactions), prices ~89% of submissions with its AI engine, and operates in 30+ countries with a long-term 10% share ambition (businessmodelcanvastemplate.com, 20 Mar 2026). Named rivals are Carwow, CarGurus, Aramis and Constellation Automotive; the cautionary comp, UK online retailer Cazoo, collapsed — a reminder that asset-heavy online used-car retail can destroy capital if unit economics don't work.

So-what: the moat narrative (data, AI pricing, pan-European logistics, captive finance lock-in) is real and confirmed externally, and low absolute share leaves a long runway. This is background that supports — but does not by itself re-rate — the stock. Positive / background.

Profit inflected up — but so did the debt funding it

Source: MarketScreener consensus financials and forecasts (AUTO1, 19 Jun 2026); 2026–2028 are estimates. Reported figures cross-checked to the FY2025 call [2].

The chart is the thesis in one frame: profitability turned the corner in 2024–25, but net debt scaled right alongside it. The bull says EBITDA keeps compounding past $580m by 2028; the bear says watch the orange bars and the missing FCF guidance.

Recent-news reference layer

Meaningful items from roughly the last three months, plus still-live events. Materiality, not age, drives inclusion.

Source: indexed news file (compiled press and analyst items) [1], supplemented by MarketScreener and FinancialReports.eu.

Specialist questions — coverage

The thesis-moving answers are promoted into the ranked findings above. The remainder sit here for reference.

Open questions a PM should still chase

The web settles most of the operational story but leaves the financial-model tail open: how the captive loan book performs through a credit cycle, what real free cash flow looks like once ABS is consolidated, and whether Autohero margin inflects in 2027. Those are the threads that decide whether this is a compounding platform or a working-capital treadmill dressed as one.

---

Figures converted from EUR at historical FX rates — see data/company.json.fx_rates. Ratios, margins, and multiples are unitless and unchanged.

Variant Perception — Where We Disagree With the Market

The honest opening: there is no directional edge here, and pretending otherwise would be the mistake. On direction the report and the market agree — AUTO1 is a genuinely good business, the wholesale Merchant engine is profitable and moated, and Autohero's margin is the swing factor. That view is the consensus: 11 Buys against 2 Holds and 1 Sell, a mean target of $35.43 versus a $27.98 spot (~27% upside), after the stock already round-tripped +62% off its spring low into the June Capital Markets Event. Being bullish on AUTO1 is not a variant view; it is the crowd.

The edge is narrower, more specific, and mostly leans the other way. It lives in two measurable gaps between what the price now pays for and what the evidence supports:

1. Pace, not direction. The crowded long has underwritten the 2027 Autohero margin inflection as an on-schedule, on-magnitude event — a ~46% step in group EBITDA from the FY2026 guide to roughly $438m. The report's evidence says management deliberately made 2026 a "hold gross-profit-per-unit flat, buy volume" year and pushed the margin step into 2027, an inflection that has slipped before [1]. We sit ~8–11% below the Street on the decisive FY2027 number.
2. Denominator, not the multiple. The market pays a debt-free platform multiple for what is increasingly a securitization-funded balance-sheet lender whose reported profit it cannot yet bank — $92m of net income against −$544m of IFRS operating cash flow, with net debt at $1.60bn behind a "no corporate debt" headline.

Both disagreements are bearish-leaning into a name where the good news is banked and the down-move is asymmetric — the −20% single-session drop on the February 2026 margin guide is the base rate. The single signal that resolves it is the next print's Autohero adjusted-EBITDA-per-unit and group GPU — not the unit headline, which is already strong.

The variant, scored

Source: analyst scoring. Consensus clarity is high because the market belief is unusually observable (14-broker coverage, a dated target distribution, a clean estimate feed, and a violent price-reaction history); variant strength is held to the low-60s because the disagreement is about pace and quality rather than direction, and direction is where the crowd and the report agree. "Weeks to next resolving print" is the estimated window to Q2/H1 2026 results (late Jul-early Aug, from the prior-year cadence); the decisive read on the 2027 setup is the Q3 print (~early Nov 2026).

The score says: a real, monetizable disagreement of medium strength, with unusually clear consensus to push against, strong evidence, and a fast clock. It is not a deep-value "the market is wrong about the company" call — those are banned and would be false here. It is a "the market is wrong about when and about what kind of asset this is" call.

What the market believes — mapped to its signals

Every row below is a market belief nailed to at least one concrete consensus signal, then translated into the testable underwriting assumption the price is actually making. Where I disagree, the disagreement is with the assumption, not the company.

Sources: ratings, targets and the FY2026/FY2027 estimate path per the analyst-estimates feed (data/estimates/analyst_estimates.json; mean target $35.43, range $23.45-$43.57) and the Web Research tab; valuation multiples per the Web Research tab; the +62% rally and price-reaction history per the Current Setup & Catalysts tab; the "no corporate debt" framing cross-read to the Q1 FY2026 results presentation [2].

One nuance the table surfaces and the price ignores: the consensus EPS the multiple rests on is already drifting down, not up — FY2026 EPS has been trimmed to $0.73 from $0.76 ninety days ago, and FY2027 from $1.11 to $1.09 (analyst-estimates feed). The market holds the "compounding justifies the multiple" belief even as the numbers it depends on are being cut. That is the seam the variant view pulls on.

The disagreement ledger

Three disagreements survived all five tests (what a consensus analyst would say; what the evidence contradicts; whether it is material; whether an observable signal resolves it on the right horizon; what would prove us wrong). They are ranked by how much each would change a PM's underwriting. Each is classified against the high-quality buckets — wrong time horizon, wrong quality of earnings/denominator, wrong liquidity/implementation — and none is a banned "high quality but undervalued" or "market too pessimistic" form.

Sources: FY2027 modelling gap and the volume-over-margin framing per the Current Setup & Catalysts tab and the Q4/FY2025 call [1]; cash-flow and leverage evidence per the Forensics tab and the cash-flow statement [3]; float, ADV and borrow-data gaps per the Short-Interest tab.

Disagreement 1 — the FY2027 step is a deferred event the price treats as a dated one

What consensus says. The FY2021-to-2025 Retail per-unit glide — from −$4,644 to −$482 — is nearly complete, so FY2027 brings the crossover, the group steps to ~$438m EBITDA (EPS ~$1.09), and the platform re-rates. Mean target $35.43 embeds it.

Where the evidence disagrees. Management itself reframed 2026 as a deliberate pause: gross-profit-per-unit held "broadly flat compared to 2025," with the $57–86m EBITDA improvement "fueled by unit growth" rather than margin [1]. That moves the entire re-rating burden onto a 2027 step that the same per-unit data shows is the hardest, last mile — Retail is still −$482 per unit against a $1,663–$2,764 segment-EBITDA-per-unit target, and the loss is sensitive to the 20–40%-growth volume mix that dilutes margin [4]. The Catalysts work models FY2027 nearer $390–407m if the inflection lands late — ~8–11% below the Street — and the estimate feed already shows FY2027 EPS being cut, not raised.

What the market must concede if we are right. That the FY2027 ~46% EBITDA step is a forecast of an undated event, not a scheduled one — and that a GPU-flat year plus a soft European cycle can push the crossover past 2027, forcing out-year EPS down rather than up.

The cleanest disconfirming signal. Two consecutive quarters of Retail adjusted EBITDA per unit stepping toward or through zero while group GPU holds. That validates the Street's schedule and kills this disagreement. Two quarters of the per-unit loss widening while GPU stays flat confirms it.

Source: FY2025 actual $232m and FY2026 guidance $287-315m (midpoint shown) per the FY2025 trading update [5]; FY2027 Street ~$438m per the analyst-estimates feed; the FY2027 "our model" $399m is the analyst's variant estimate carried from the Current Setup & Catalysts tab.

Disagreement 2 — a lender priced as a platform; profit the market cannot yet bank

What consensus says. AUTO1 is a debt-free, self-funding platform; the ABS facilities are matched, non-recourse working-capital plumbing, so the deeply negative IFRS operating cash flow is an accounting quirk and the platform multiple is fair.

Where the evidence disagrees. The audited record shows reported profit and cash moving in opposite directions: $92m of net income against −$544m of IFRS operating cash flow in FY2025, with the cash position held only because +$560m of financing draws refilled the burn — a pattern structural since FY2022 [3]. Net debt is $1.60bn (D/E 2.26x, from 0.92x in FY2022), and the "No Corporate Debt" slide that anchors the bull narrative excludes $1.05bn of inventory-ABS liabilities that were already 87% drawn at Q1 2026 [2]. The Forensics tab grades the gap Elevated (57/100) precisely because the company's headline "AUTO1 Cash Flow" of +$232m sits $777m above the −$544m IFRS figure. And the company's own FY2022 report names liquidity "the most relevant potential financial risk," conceding it needs continued bank and capital-market access until it reaches positive operating cash flow [6]. Layered on top, the captive-finance book — receivables up ~50% in 2025 — already produced a $13.9m credit charge management called a one-off, on a portfolio young enough that a "one-off" is hard to distinguish from an early read on reserve adequacy [7].

What the market must concede if we are right. That a meaningful and growing slice of reported profit is credit-derived and funding-dependent, that the right denominator is $1.60bn of net debt rather than zero, and that the multiple should carry a discount for reliance on continuously open, cheap ABS.

The cleanest disconfirming signal. Group IFRS operating cash flow turning positive without fresh ABS draws for two or more quarters — the company's own "self-funding" claim made real. That, plus a captive-finance book that seasons without a second charge, refutes this disagreement.

Source: reported net income and IFRS operating cash flow, FY2022-FY2025, per the Forensics tab and the cash-flow statement [3].

Disagreement 3 — analytically right, institutionally hard

What consensus says. A $6.1bn cap covered by 14 brokers is a liquid mid-cap that can be sized and exited at will.

Where the evidence disagrees. Only 23.2% of the share capital was placed at IPO; founders (~21%) and SoftBank (~14.8%) sit on a concentrated register, the true free float is thin, and average daily turnover is only ~$16m (~0.75m shares). The Short-Interest tab is explicit that no aggregate short-interest, short-sale-volume or borrow data exists for this Xetra name — so a PM cannot currently size the cost or feasibility of the short leg, and a sponsor block sale into a thin float is an asymmetric-down technical the analytics ignore. This is the one disagreement that is not about being more bearish — it is about a thesis being correct and still difficult to express at institutional size.

What the market must concede if we are right. That the implementation constraint — float, borrow, overhang — is a real input to expected return for this name, not a footnote.

The cleanest disconfirming signal. BaFin/Bundesanzeiger net-short publications above 0.5%, an announced placement or block, or realized impact-cost on the tape that shows the float absorbs size cleanly.

The evidence layer — what a PM can audit fast

These are the items that actually move the probability of the variant views, each paired with the consensus read, our read, and — critically — its fragility, the thing that could make the evidence misleading.

Sources: estimate path and revisions per the analyst-estimates feed and Web Research tab; the 2026 GPU framing cross-read to the Q4/FY2025 call [1]; cash-flow and ABS-draw evidence per the Forensics tab, the cash-flow statement [3] and the Q1 FY2026 presentation [2]; the $13.9m credit charge per the Q1 FY2026 call [7]; float and borrow gaps per the Short-Interest tab.

How this resolves — observable signals only

Each signal below is observable in a filing, an earnings call, the tape, an estimate revision, or a regulatory docket — never "better execution" or "time will tell." The top two resolve durable thesis variables; the rest inform.

Sources: the resolving-evidence definitions track the long-term thesis spine - Retail per-unit economics [4], the cash-flow/ABS framing [3] [2], and the captive-credit charge [7] - per the Long-Term Thesis, Forensics and Current Setup & Catalysts tabs; estimate path per the analyst-estimates feed.

Red team — what would kill these views before the market does

This is written to break the variant, not to protect it.

The funding disagreement has its own kill switch. The capex intensity is genuinely ~0.3% of revenue, the trading (car) business is substantially self-funded by matched non-recourse ABS, and a fresh consumer-loan securitization (FinanceHero-2, $278m, Sep 2025) proves the ABS market is open and pricing AUTO1's paper. If FY2026 operating cash flow turns positive on its own as inventory growth normalizes, the "lender priced as a platform" framing loses most of its force — and the company's "self-funding" claim becomes the more accurate description than ours. The audited statements are unqualified, with no restatement, probe, or auditor flag in the record; this is a quality-of-earnings discount, not a fraud flag, and a clean season of cash conversion removes it. Finally, the tradability disagreement is the most easily neutralized of the three: if no block sale materializes and the float absorbs flow, it stays a footnote rather than a cost.

The intellectually honest read is that all three disagreements are bearish-leaning into a name whose business has genuinely inflected — which is exactly the setup that punishes a stubborn short. The variant view is not "AUTO1 is a bad business." It is "the price has underwritten the best version of the next 18 months, on schedule, on a platform multiple, and the evidence says the schedule and the denominator are both more uncertain than the multiple admits."

The one signal to watch first

---

Short Interest & Thesis — AUTO1 Group SE (AG1)

Figures converted from euros (EUR) at historical FX rates — see data/company.json.fx_rates for the rate table. Ratios, margins, and multiples are unitless and unchanged.

Bottom line

Reported short interest is not decision-useful for AG1 today: no official aggregate short-interest series, no daily short-sale-volume series, no public net-short threshold disclosures, and no borrow/securities-lending indicators were staged for this German (Xetra) name. So this page cannot tell a PM whether the position is crowded, hard to borrow, or set up for a squeeze — and it says so plainly rather than dressing up a number that does not exist.

What is decision-useful is the thesis-risk side, and it is built entirely from the primary record. There is no credible public short-seller report or activist campaign in the staged evidence, but there is one real, checkable lever a short would lean on: AG1 now reports positive net income ($92m in FY2025) while its IFRS operating cash flow ran deeply negative (about -$544m in FY2025), because the asset-backed-securitisation (ABS)-funded build of inventory and captive-finance receivables shows up as an operating outflow. Management addresses this directly in its 2026 Capital Markets material. The strongest evidence is the corpus itself (risk factors, the ABS capital structure, management's own cash framing); the biggest gap is the complete absence of positioning and borrow data.

Evidence quality — what exists, what does not

Source: short-interest data availability per staged data/short_interest/ manifest (reported short interest, short-sale volume, disclosures, borrow — all staged as zero rows); EU SSR channel per the staged source manifest.

The staging manifest is explicit: official_reported_short_interest_available is false, every position/volume/disclosure/borrow table is empty, and ADV is null. The only official source the data step even considered for this name was the EU Short Selling Regulation threshold-disclosure channel — which, by construction, captures only holder-level positions above 0.5% of capital and is not a complete aggregate short interest. Treat any later claim that "AG1 short interest is X" with suspicion unless it cites a dated official position report.

Crowding versus liquidity — context, not a verdict

Without shares-short or days-to-cover, AG1 cannot honestly be called "crowded" or "un-crowded." What the price/volume feed does give is the liquidity backdrop a future short-interest read would be judged against.

Source: price, volume and shares-outstanding from the staged numeric feed (data/prices/, data/financials/); ADV and market cap derived from that feed. Last close $27.98 (19 Jun 2026); ~218.8m shares outstanding.

At roughly $28 the stock trades about 36% below its $43.04 February 2021 IPO price, and its ~$6.1bn market cap is well under the ~$9.0bn implied at listing [1]. Average daily turnover of roughly $16m (about 0.75m shares) is moderate for a $6bn name: even a hypothetical short position equal to a few days of float would take many sessions to cover, so if a meaningful short ever showed up, liquidity would amplify rather than cushion a cover. That is a conditional statement — there is no staged short position to apply it to.

One structural caveat on float: only 23.2% of share capital was placed at IPO, with pre-IPO holders (and cornerstones Sequoia Capital and Lone Pine) holding the balance under a 180-day lock-up at the time [1]. A concentrated register tends to mean thinner true free float and thinner lendable supply — a reason borrow data, if it existed, would matter here.

The short-thesis ledger — built from the primary record

There is no staged short-seller report. But a disciplined short case for AG1 would not need one; it would lean on disclosures the company itself makes. Each row below pairs the lever a short would press with what the company's own filings show and how management frames it.

Sources: profit-vs-cash and ABS framing — Capital Markets Event presentation (17 Jun 2026) [8] and [9]; ABS programmes, equity ratio, liquidity-risk and going-concern language — FY2022 Annual Report [2], [4], [5], [6].

The one lever that matters: reported profit versus cash flow

This is the crux of any AG1 short thesis. The numbers are stark.

Source: reported financials, FY2022–FY2025 (data/financials/income.json, cash_flow.json); accumulated deficit and ABS debt cross-read to the FY2022 Annual Report [3].

The widening green-versus-red gap is exactly what a forensic short would screen on: as the model scales, reported earnings improve but IFRS operating cash flow gets more negative, not less, and cash on the balance sheet fell to about $289m at FY2025 from $387m a year earlier. A naive reading says "this company earns profit it cannot collect."

Management's rebuttal is specific and disclosed, not hand-waving: under IFRS the increase in car inventory and captive-finance receivables is booked as an operating cash outflow, while the ABS funding that finances those very assets is booked in financing cash flow — so the operating line overstates the real cash strain [8]. On the company's funded view, inventory growth and captive-finance growth are largely offset by their matched ABS funding.

Source: Capital Markets Event presentation, "AUTO1 has Strong Track Record in Generating Cash" cash bridge (17 Jun 2026) [8].

The institutional read: this is a real unresolved debate, not a fraud flag. The short lever is genuine (IFRS operating cash flow is deeply negative and the model depends on continuously rolling ABS funding to grow), and the rebuttal is genuine too (the funding is matched, non-recourse, and the net exposure is far smaller than the headline outflow). The thesis risk therefore lives in credit/funding-market access, not in an accounting deception — and that is where a PM should focus diligence.

The capital structure a short would attack

Source: reported balance sheet, FY2025 (data/financials/balance_sheet.json); ABS-programme structure and equity-ratio framing per FY2022 Annual Report [2], [4].

AG1's only material financial debt sits in two non-recourse ABS programmes run through dedicated financing companies (AUTO1 Funding B.V. and Autohero Funding 1 B.V.): an inventory ABS facility ($485m drawn at FY2022) secured by the used-car inventory with no further recourse to the group, plus a consumer-loan ABS facility refinancing Autohero instalment receivables [2]. The company itself flags liquidity risk as "the most relevant potential financial risk" and states it will need continued access to banks and capital markets to fund inventory and car loans until it reaches positive operating cash flow [5]. That is the cleanest disclosed bear lever on the page: the model works only as long as the ABS market stays open and cheap.

The mitigants are also in the record. Equity was $731m at FY2022 with a 40.6% equity ratio (down from 56.7%, driven by accumulated losses) [4], and neither management nor the auditor framework identified any risk endangering the company as a going concern — there is no modified or qualified opinion in the readable record [6]. A short thesis built on "going concern" would be contradicted by the primary record; a short thesis built on "ABS-funding fragility in a credit shock" would not.

Where the turnaround cuts against the bears

The disclosed operating trajectory is the single biggest problem for a short. Group adjusted EBITDA swung from -$177m in FY2022 to +$232m in FY2025 (a 2.4% margin), and the consolidated loss of $262m in FY2022 has since flipped to reported net profit [3] [7]. A short pressing the cash-flow lever is fighting an improving P&L and a structurally widening unit economics gap (rising gross profit per unit). That asymmetry — improving fundamentals against a funding-structure worry — is precisely why positioning data would be valuable here, and precisely what is missing.

Public net-short disclosure regime — the channel to watch

Source: EU SSR disclosure channel per the staged short-interest source manifest (data/short_interest/source_manifest.json); zero disclosures staged.

AG1 trades on Xetra and falls under the EU SSR. The right place to look for named short holders is the German Federal Gazette (Bundesanzeiger) / BaFin net-short-position publications above the 0.5% threshold. None were staged for this run. The institutional caution: even when populated, that channel shows only large individual positions above the threshold — it is a floor on disclosed shorting, never the full aggregate. Do not let a "no disclosures above 0.5%" reading be mistaken for "no short interest."

Market setup

With no positioning or borrow data, there is no squeeze or forced-de-risking signal to read. The setup risk that is visible is event-driven, not flow-driven: the contested item is the cash-flow narrative, and the catalysts that move it are credit-market conditions (ABS spreads, facility renewals) and each quarterly cash bridge. A short pressing the funding lever is exposed to exactly the upside the bulls own — continued margin expansion and the company's funded-cash-flow framing gaining acceptance — while a long is exposed to a credit-market shock that widens ABS spreads or constrains advance rates. The absence of borrow data means a PM cannot currently size the cost or feasibility of expressing either side via the short leg.

Limitations

This page is deliberately thin on positioning because the positioning data is absent. To be explicit: there is no staged aggregate short interest, no short-sale volume, no public net-short disclosure, no borrow fee/utilisation/lendable-supply data, and no source-dated peer short comparison. ADV is derived from the price feed, not provided. The thesis-risk analysis is grounded in the readable corpus — but note that AG1's own FY2023–FY2025 annual reports and all interim results PDFs in this corpus failed to parse (AES-encrypted), so the most recent disclosed risk-factor and notes language could not be read directly; recent figures lean on the results/Capital Markets presentations and the numeric feed instead. The cleanest next step is to pull the official BaFin/Bundesanzeiger net-short publications and a borrow-cost read before forming any positioning view.