Financial Shenanigans

The Forensic Verdict

Forensic Risk Score: 18 / 100 — Clean. Auto Trader's reported numbers behave the way a high-margin, low-capex marketplace should: cash flow runs ahead of earnings every year, receivables grow slower than revenue, capex is a rounding error, and the income statement reconciles cleanly to the cash statement without working-capital lifelines. Two items deserve watching — the £11.1m Autorama deferred-consideration charge that flowed through "adjusted" measures in FY24, and the 73.8% intangibles share of the balance sheet — but neither is thesis-changing. The single piece of evidence that would most change the grade is a goodwill or Vanarama-brand impairment in FY26: the brand's useful life was already cut to five years in October 2023, and a further write-down would suggest the £148m Autorama purchase was overpaid in a way prior disclosures have not yet recognised.

Breeding Ground

The incentive structure points at growth metrics, but pay flexes down when targets are missed and the auditor and board behave normally. Two structural items keep this from being a perfect score: bonus and Performance Share Plan ("PSP") metrics are weighted to operating profit and revenue growth — exactly the metrics most vulnerable to aggressive recognition or cost capitalisation — and KPMG is in year eight of a mandatory ten-year audit-tender cycle. Neither has produced visible distortion.

The most material breeding-ground signal is the pay metric mix. PSP awards vest 70% on operating-profit growth and 20% on revenue growth — both metrics the company has direct accounting control over. Two things keep this from translating into earnings management. First, the FY25 outcome shows real bite: a slight miss on both PSP metrics, the CEO's single-figure pay dropped 26%, and the bonus paid at 43% of maximum versus 92% the year before. Second, capex and capitalised-cost behaviour is restrained — capex/revenue is 0.7%, capex/depreciation is 19% — meaning there is no obvious lever being pulled to flatter operating profit. The third leg, CEO skin in the game, is unusually strong for a UK plc: Coe's stake is c.36x base salary versus a Code guideline of 200%.

The KPMG tenure is on schedule, not late: the firm was appointed at IPO in 2015 and the first mandatory tender will run in FY26 ahead of the FY27 audit. That is the normal cycle, not a defensive rotation.

Earnings Quality

Earnings look earned, not engineered. The most powerful test for a subscription marketplace is whether revenue growth is matched by cash collection. Auto Trader passes it cleanly: receivables grew 1.7% in FY25 against revenue growth of 5.3%, and DSO has been within a 50–53 day band for four consecutive years. Operating margin expansion (54.0% FY23 to 61.1% FY24 to 62.7% FY25) is supported by recurring subscription pricing power and Autorama loss reduction (£8.8m to £4.3m), not by capitalisation gimmicks.

Revenue vs receivables — the cleanest test

The single year where receivables grew faster than revenue was FY24 (+14.3% vs +14.1%) — and even there the gap is 20 basis points. The structural pattern is the opposite: in FY22 (post-COVID rebound) revenue +64.6% pulled receivables only +10.6%, and in FY25 a slowdown to +5.3% revenue growth was matched by near-flat receivables. There is no quarter-end stretch, no bill-and-hold, and no contract-asset balance flagged in disclosure.

Margin expansion — sourced from operating leverage

The FY23 dip was Autorama dilution — a £148m acquisition of a loss-making vehicle-leasing business consolidated mid-year. Operating margin compressed from 70.2% to 55.5%, then rebuilt to 62.7% as Autorama losses narrowed. Crucially, net margin and EBITDA margin moved in the same direction every year — there is no disconnect between accrual earnings and underlying cash profitability.

Capex versus depreciation — capitalisation discipline

This is one of the cleanest capitalisation patterns in any FTSE 100 name. Capex has run between £1.3m and £4.0m over five years on a revenue base that grew from £263m to £601m. D&A is dominated by amortisation of acquired Autorama intangibles (Vanarama brand £12.9m FY25), not by software or capitalised internal-use assets that could mask operating expenses. If Auto Trader were aggressively capitalising development costs, capex would be visibly higher than depreciation — instead it is a fifth of D&A. This is the single strongest piece of negative evidence against earnings inflation.

Cash Flow Quality

Operating cash flow is structurally durable: built from subscription collections, not from working-capital lifelines or financing dressed as operations. Cash conversion has run above 1.0x net income in four of the last five years, with FCF margin holding above 49% in FY24–FY25. There is no factoring disclosure, no supplier-finance programme, and no securitisation. The only cash-flow item that demands scrutiny is the FY23 acquisition of Autorama, which absorbed £152.3m of investing cash flow — and even after that, FCF after acquisitions was £112m positive in FY23 and has been £284m and £301m in the two years since.

CFO vs Net Income — five years of >1.0x conversion

The pattern is the high-quality version of the chart: cash flow above net income every year except FY21 (which was COVID-distorted and had no bonus accrual). The 3-year average CFO/NI of 1.11x and FCF/NI of 1.10x are exactly what a high-margin, low-capex marketplace should deliver, and they reconcile to the income statement without "adjusted" cash-flow definitions. The accrual ratio (NI minus CFO, divided by average total assets) is negative every year in the FY22–FY25 window — the cleanest possible reading.

FCF after acquisitions — the deal-adjusted view

Auto Trader is not a serial acquirer — Autorama in FY23 is the only material deal in five years, and the post-deal FCF of £112m still funded ordinary dividends. In FY24 and FY25 there were no acquisitions at all, so headline FCF and acquisition-adjusted FCF are identical. This rules out the most common forensic distortion at acquisitive UK compounders: hiding integration spend in investing rather than operating.

Working-capital contribution — small and stable

The £22m–£33m gap between CFO and net income across FY22–FY25 is dominated by depreciation and amortisation (£7m–£21m) plus share-based-compensation add-back (£3m–£10m) and modest working-capital movements. There is no year where a working-capital release accounts for the bulk of cash-flow strength — the lifeline test fails to find a lifeline because the underlying business does not need one.

Metric Hygiene

The non-GAAP gap is small, the definitions are stable, and the metric Auto Trader pushes hardest — operating profit — is a GAAP line. Two judgment calls are worth flagging. First, FY24 adjusted EBITDA excluded an £11.1m deferred-consideration charge tied to Autorama; that adjustment goes away in FY25, mechanically inflating "growth" comparisons. Second, the reduction of the Vanarama brand's useful life to five years in October 2023 accelerated amortisation from £10.0m (FY24) to £12.9m (FY25), with a further step expected in FY26.

The gap between statutory operating profit and adjusted EBITDA peaked in FY23 at £31m (Autorama acquisition costs and brand amortisation were freshest) and has compressed to £17m by FY25. This is the right shape: as integration matures, the "adjustments" should shrink. Reverse the chart and the gap is widening — that would be the warning sign. It is not.

The one disclosure-quality issue worth keeping on the watchlist is the absence of a separate impairment review disclosure for the Vanarama brand. The October 2023 useful-life reduction implies management identified a shorter economic life than originally booked at acquisition, but no separate impairment charge accompanied it. If Autorama loss-narrowing stalls, an impairment becomes the cleaner explanation.

What to Underwrite Next

Accounting risk does not warrant a valuation haircut or a position-sizing limiter on this name. The forensic profile is consistent with a high-quality, low-capex, asset-light marketplace whose reported numbers approximate economic reality. What it should affect is the watch list — there are four specific items that, if they move the wrong way, would push the grade from Clean to Watch or Elevated.

1. Vanarama brand or Autorama goodwill impairment in FY26 or FY27. The brand's useful life was already cut to 5 years in October 2023 without an associated impairment charge. If the FY26 accounts add an impairment, that is a tell that the £148m purchase price now requires retroactive accounting recognition. Track the intangibles roll-forward in note 12 of the AR.

2. DSO drift above 55 days. Current band is 50–53 days. A move to 55 or beyond would flag credit-quality deterioration in the dealer base or a stretching of payment terms — early indicator of revenue-recognition pressure.

3. PSP metric mix. The next remuneration policy refresh is due in FY26. If the weighting shifts further toward growth metrics or introduces an "adjusted" earnings target without a TSR underpin, the breeding-ground risk increases. Currently 70% operating profit growth, 20% revenue growth, 10% carbon — manageable.

4. CMA "fake reviews" probe outcome (opened March 2026). Forensic-relevant if it forces re-presentation of platform metrics (reviews, ratings) that feed retailer-facing KPIs. A remedy package would not directly hit the income statement but could invalidate certain ARPR-product line attributions in disclosure.

What would upgrade to a 10–15 score (clean-cleaner): KPMG re-tendered or replaced cleanly in FY27, two more years of FCF/NI above 1.05x, intangibles share of total assets falls below 70% as Vanarama amortises, and the FY26 PSP outturn shows continued discipline on stretch targets.

What would downgrade to 35–45 (Watch–Elevated): any goodwill impairment, a year of receivables growth materially exceeding revenue growth, capex stepping up to over 2% of revenue with no corresponding revenue acceleration, or an audit committee finding flagged in the next AR.

Bottom line for the PM. The accounting is not a footnote risk that needs separate underwriting and it is not a thesis breaker. It is genuinely clean. The decision-relevant risks at this name sit elsewhere — competitive pressure from Cazoo-era new entrants, the CMA probe, the Autorama integration arc, and the maturing UK used-car ad market. None of those are accounting risks. The reported numbers can be relied on for valuation work without a forensic discount.