TLDR:
Two finance people, neither of them an accountant, have raised €30m to build a top-10 French accounting firm from scratch — four firms and ~180 staff in eighteen months. Numeris is the clearest European test yet of a thesis the Americans have already proved.
Accounting ticks every box a consolidator wants: fragmented, recurring, cash-generative, and full of owners heading for the exit with no succession plan. The proof it pays — Citrin Cooperman sold for ~$2bn in 2025, about 4x what it fetched three years earlier.
So is it a genuine compounding machine or just an expensive way to buy revenue? Below: the economics, the proof, and the traps — with Numeris as the test case.
Why is everyone buying up accountants?
In January 2025, Blackstone took a majority stake in Citrin Cooperman, a top-20 US accounting firm, in a deal valuing it at around $2bn. The seller, New Mountain Capital, had paid roughly $500m a little over three years earlier — and in between had bolted on a string of firms and taken revenue from $350m to $850m.
A 4x on an accounting firm, in three years. Not, on the face of it, the asset class you'd expect it from.
Citrin isn't a fluke — it's the clearest sign yet that one of the staidest professions going has become a serial acquirer's market, and not only in the US.
Accounting is not a glamorous business. It is, however, exactly what a serial acquirer looks for, because it ticks every box at once:
A long tail, and no leader. France alone has more than 20,000 firms turning over around €20bn, with no single player above ~5% share and two-thirds employing fewer than ten people — one of the most fragmented markets in Europe. The US is bigger again, at close to 90,000 firms. Fragmentation is the raw material of consolidation.
Boringly essential. Tax and accounts aren't optional, and once a business picks a firm it rarely leaves: 77% of French companies use an accountant, more than half have used the same one for over a decade, and consolidators report client churn below 5% and revenue retention above 100%. The revenue simply recurs.
A cash machine. Past the people, a firm needs almost no capital, so EBITDA is effectively cash flow. Dull — and precisely what you want to put debt against.
A generation heading for the door. Around three-quarters of US accountants were at or near retirement age by 2020, the average is in the early fifties, and too few juniors are replacing them — which, for an owner with no successor, makes it a buyer's market for practices. France is ageing on the same curve. Motivated sellers, sensible prices.
A regulated moat — with a catch. Not anyone can practise: the work is licensed, and in France the profession is governed by the Ordre des experts-comptables. That keeps competition out — but the same ownership rules dictate how outside capital gets in, which turns out to matter enormously. (More on that later.)
None of this is hypothetical. Private equity only moved into American accounting in 2021, but in the few years since it has worked through the profession's upper ranks — and Citrin shows what the exits can look like. What's new now is that the playbook has crossed the Atlantic — European PE deals in the sector have gone from a handful a year before 2022 to roughly 200 in 2024 — and that founders, not only funds, are running it.
The questions worth asking are whether it travels, whether it pays at European prices, and who comes unstuck trying. Numeris is the cleanest case to think it through.
So where does the 4x come from?
Part of it is the same financial trick behind every roll-up: small firms trade at a fraction of the multiple a large platform commands, so each acquisition creates value on the spread alone. I set out that arithmetic in the technical-services piece and won't repeat it — but it works identically here, which is why a firm picked up at a high-single-digit multiple can sit inside a group that changes hands at 15x.
Arbitrage alone, though, is a financial trick, not a business. What makes accounting genuinely interesting is what you can do to the firms once you own them:
Move up the value chain — but keep it recurring. Commoditised bookkeeping is the base of the trade, and it's increasingly automated and outsourced away. The value sits in the recurring, relationship-based work above it — tax, compliance, outsourced finance, payroll — the low-risk, sticky band in the chart below, and where the consolidators concentrate. In France, routine accounting is still about half of sector revenue but is forecast to fall to ~37% by 2030, while advisory grows 10–12% a year. The trap is mistaking that prize for high-end project consulting — lucrative in a good year, lumpy across a cycle. The money is in the work clients need every year, not the work they buy once.
Sell more to the clients you already have. The accountant sees the owner's whole financial picture, which makes cross-sell unusually easy. Citrin's revenue is already over 50% tax, 25% advisory and only 20% audit — a long way from just keeping the books. Every service added deepens the relationship and lifts switching costs at the same time.
Use scarce labour two ways. A shortage of qualified accountants is a headache for everyone — but for a consolidator it's pricing power, because clients can't easily shop around. American groups go further and offshore the routine work: Citrin took its India team from a handful to over 350 in seven years. That lever leans on language, though — English bookkeeping moves easily to India or the Philippines; French or German compliance work does not. Whether European consolidators can match American labour costs, and American margins, stays open.
Then the catch — the one that decides whether any of this works. Every lever above depends on keeping the people. Clients are sticky, so most stay — but the relationships are held partly by the firm and partly by individual partners, and a partner who leaves can take the ones they personally own. That's why how a deal is structured — how much equity the seller keeps, and for how long — matters at least as much as the price. Numeris's whole model is an answer to that problem.
So who's doing this in Europe?
Numeris was founded in 2024 by two ex-PE financiers, neither of them an accountant. By late 2025 it had raised €30m from Strada Partners, a Belgian fund, and built four firms, ~180 employees and around €25m of revenue from a standing start. The target: a top-10 French firm within ten years, €200m of revenue inside five.
The model is the interesting part, because it's their answer to the retention problem. Numeris sells a "third way" between staying independent and being swallowed: firms keep their brand and full professional autonomy while group-level support — finance, HR, IT, recruitment, an in-house academy, internal software — does the heavy lifting.
Two things stand out. It's pointedly not an "AI roll-up" — technology is a lever, not the thesis — and its senior adviser ran the Paris Ordre, which counts in a profession that polices who may own it. What it lacks is a track record: eighteen months old, nothing sold. For evidence the model works, look at the firms that have run it for years.
Does it actually work?
Numeris is unproven. The model isn't — and the clearest place to watch it work is Kelly Partners, an Australian accounting roll-up listed since 2017 that runs almost exactly the structure Numeris is copying: autonomous partner-run firms, 51% owned by a listed parent and 49% by the operating partner, with acquisition debt held at the firm and non-recourse to the group. It is the Numeris pitch with two decades of receipts.
The receipts are good. Kelly has compounded revenue at around 30% a year to A$134.6m, overwhelmingly by acquisition, and the firms it owns are profitable: established practices run at ~31% EBITDA margins against ~25% for subscale ones. Buy at a single-digit multiple, integrate, push the margin toward thirty, repeat — that's the whole machine, running in public.
The exits prove the cash gets out. Citrin's $500m-to-$2bn in three years is the headline, and it isn't alone: EisnerAmper, the other 2021 pioneer, recapitalised in 2026 rather than sell — a sign its backers think there's more to come. In the US, CBIZ has run the same playbook in public-company daylight for decades, using an attest/non-attest split to get round the rule that non-accountants can't own a CPA firm.
That rule is the thread back to Europe. Simago, a French radiology roll-up, sold for ~€525m at a 10x return in under three years by exploiting a cap on non-professional ownership of medical practices — the same species of constraint that runs through accounting. EU rules require statutory audit firms to be majority-owned by qualified auditors, so outside money steers toward the work that isn't statutory audit — accounting, tax and advisory — where an investor can take majority control. That's exactly Numeris's patch, and why a fund can hold the upper hand without breaking the rules. The ownership rules don't block the roll-up; they decide which corner of it you can own. That is the bet Numeris is making.
Where does it break?
The bull case is real and largely proven. So is the bear case — and most of it hides inside the same Kelly numbers that make the model look good.
The growth is bought, not earned. Strip out acquisitions and Kelly's underlying business grows at about 4.5% a year; the rest is bolt-ons. A consolidator that stops buying stops growing — and buying only gets harder as the field fills with funds chasing the same firms and prices rise. The machine runs on a steady supply of cheap deals and cheap debt; remove either and it stalls.
A people business — but a sticky one. Clients don't move easily; switching accountant is a hassle, which is why churn runs under 5%. Most of the relationship sits with the firm — the records, the systems, the continuity — but part of it sits with the individual partner, who can take the clients loyal to them personally. So the risk isn't mass client flight; it's a key partner walking, or coasting once the earn-out is paid. The retained-equity structure exists precisely to stop that.
The regulator is watching. Loading a public-interest profession with profit-driven owners makes oversight bodies uneasy: the US PCAOB is reviewing how PE ownership affects auditor independence, the global ethics board has issued an alert, and in France the Ordre controls who may own a firm at all. None of this stops the roll-up — but a tightening of the ownership rules is the tail risk that would hit every player at once.
None of this is fatal — it's the difference between a disciplined compounder and an expensive way to buy revenue. The model works; the execution is where it's won or lost.
So, is it worth doing?
Strip away the novelty and the picture is unusually clear. The economics are real: fragmented supply, recurring revenue, cash that converts, and exits that have paid out at 4x. The model is known — Kelly Partners has run it in public for nearly twenty years. And the risks are just as well understood: most of the growth is bought rather than earned, retention is everything, and the regulator could yet change the rules.
What that combination rewards is operators, not financial engineers. The multiple arbitrage is just arithmetic — anyone can do it on a spreadsheet. The hard part, the part that decides whether you compound at 30% or quietly destroy value, is integrating partner-led firms and keeping them committed once the cheque clears. That is an operating problem, not a finance one.
And the field is filling fast — but unevenly. Globally private equity put $6.3bn into accountancy in 2024, the most in a decade, yet the map is lopsided: the UK accounts for 41% of Europe's PE accounting deals, France just 4%. For anyone eyeing the Continent, that gap is the opportunity — most of Europe is barely touched.
So if you're weighing your own roll-up, the lesson isn't to outbid Cinven or Blackstone for the trophy firms. It's that the market is so fragmented, that the long tail sits wide open at sensible prices — and the binding constraint isn't capital, of which there is plenty, but the operating skill to keep partners and integrate firms — the genuinely hard part, and where the next decade of winners will be decided. The thesis is proven, the sellers are lining up, and most of the Continent is still in play. The only real question is who's prepared to do the unglamorous work.
Disclaimer: Stripped back to just the facts and a few opinions. Views are my own, sources public, none of it financial advice. Best consult a professional before doing anything with real money.