What Buyers Actually Pay For: The Five Levers Behind a Service Business's Multiple

Sit on the buyer's side of the table for a moment. An acquirer looking at your consultancy, agency, or training business isn't paying for what you earned last year. They're pricing how confident they can be about what the business will earn after you hand over the keys. Every diligence question, every spreadsheet request, every awkward interview with your team is an attempt to measure that confidence.

This is why two businesses with identical top lines can sell at wildly different prices. Picture two service platforms, each doing EUR 2 million in annual revenue with roughly 150 certified practitioners, operating in similar markets with similar methodologies. One changes hands at 3x revenue — EUR 6 million. The other commands 10x — EUR 20 million. The EUR 14 million gap has nothing to do with performance. It's architecture.

John Warrillow has built a career around this exact puzzle. Across Built to Sell and The Automatic Customer, he keeps arriving at the same conclusion: a business's value is not a function of how much revenue it generates, but of the structural traits that make future revenue stable, defensible, and capable of growing without the founder in the room.

And here's the part founders miss: this applies even if selling is the last thing on your mind. The traits that earn a premium from a buyer are the same traits that let a business scale, survive your absence, and produce wealth instead of merely paying you a salary. Warrillow's advice to build as though you'll sell isn't about exiting — it's about forcing the disciplines that make the business excellent either way.

Five structural levers set the multiple. Below, each one is framed the way an acquirer frames it — as a question they need answered before they write a number on a term sheet.

The Dealbreaker Question: What Happens When You Leave?

Start with the lever that can kill a deal outright. Warrillow is blunt about it: "Nobody buys a company that can't function without its owner." Before a buyer cares about your growth rate or your margins, they want to know whether the thing they're buying still exists once you're gone.

Note what this question is really probing. It's not asking whether you're a capable operator — most founders of expertise businesses are. It's asking whether your judgment has been transferred into systems, processes, and people who act on it without you. If it hasn't, the buyer isn't acquiring a business; they're acquiring you. And the moment your earnout ends — or you burn out, or you get bored — the asset they paid for walks out the front door.

Independence maps to value in a fairly direct way:

• A week without you and things break: expect 1-2x revenue at best. Whatever it says on your LinkedIn profile, this is a job dressed up as a company. The value evaporates when the founder does.
• A month without you: the 3-5x range opens up. Systems exist, but strategic calls, key client relationships, and quality oversight still route through the founder.
• Three months or more without you: 5-12x territory. Governance bodies make decisions, practitioners generate their own deal flow, and quality mechanisms run on their own. The founder adds value when present but isn't load-bearing.

This is where Mike Michalowicz's Four-Week Vacation test earns its keep. It reads like a lifestyle exercise, but it's really a valuation exercise. Each time you take the month off and the business holds, you've produced evidence of founder independence — and that evidence translates directly into money when a buyer is assessing risk.

The Certainty Question: How Much of Next Year Is Already Sold?

From Project Multiples to Platform Multiples

If founder independence is the dealbreaker, recurring revenue is the lever with the largest range of motion. A business where 90% of revenue recurs trades at 3 to 8 times the multiple of an otherwise similar business where 90% comes from projects. On a EUR 2 million business, that's the gap between a 2x outcome (EUR 4 million) and a 10x outcome (EUR 20 million). Nothing about the work changed — only the structure of how it's sold.

The logic is simple. A project business resets to zero every quarter: each engagement requires its own negotiation, its own proposal, its own close. A subscription business — annual certification fees, platform subscriptions, data access licenses — opens each year with 80-90% of the prior year's revenue already committed. A buyer pricing the future will pay far more for the second pattern, which is why Warrillow built an entire methodology around engineering it.

"A project pipeline tells a buyer what you might win. A subscription base tells them what they already own."

For a methodology business making this transition, the recurring percentage typically climbs in stages:

• Year 1 — 20-40% recurring. Direct services are sunsetting while the founding cohort comes in on free terms. The number looks low because you're laying the base while still delivering the old way.
• Year 2 — 60-80% recurring. The founding cohort converts to paid, new cohorts join on annual subscriptions, platform fees go live, and direct services are nearly gone.
• Year 3 and beyond — 80-95% recurring. Almost the entire base is annual subscriptions, licensing, and platform usage. Growth now comes from cohort expansion and pricing, not from selling new projects.

The threshold that matters is 80%. Cross it, and your business stops being priced like professional services (2-4x revenue) and starts being priced like a subscription or platform company (5-12x revenue). No other single structural change moves the number that far.

One accelerant worth naming: bill annually, not monthly. Annual upfront billing means cash lands before delivery happens, which pushes your Cash Conversion Cycle negative — and a negative cash conversion cycle tells a buyer the business finances its own growth.

The Moat Question: What Would It Cost to Rebuild You?

Recurring revenue tells a buyer the income is stable today. The next question is whether someone else can take it away tomorrow. In a service platform, the honest answer hinges on one thing: the network. Your methodology can be imitated. Your technology can be rebuilt. What resists copying is the web of practitioners, client relationships, referral patterns, accumulated assessment data, community norms, and brand trust that took years to form.

So run the test a buyer runs. If a competitor could poach your top 10 practitioners and walk away with 80% of your value, you own a people business wearing a platform costume — and the price will reflect that fragility. If replicating you means rebuilding the whole ecosystem from nothing, you own a moat worth paying for.

The moat gets its depth from three reinforcing network effects:

• Same-side effects. Practitioners make each other more valuable — through referrals, shared learning, and complementary specialisations. Each new member raises the worth of membership for everyone already inside.
• Cross-side effects. More practitioners mean better coverage, which draws more clients; more clients mean more deal flow, which draws more practitioners. Past critical mass, the loop feeds itself.
• Data network effects. Every assessment enriches the benchmarking database, and a richer database makes each subsequent assessment more valuable. This is the deepest layer — 5,000 assessments spanning 15 industry segments cannot be reproduced without rebuilding the ecosystem that generated them.

Buyers compress all of this into a build-versus-buy estimate: what would it cost us to create this ourselves? "Three years and EUR 5 million, with no guarantee it works" is the answer that earns a premium. "Six months with a decent sales team" is the answer that doesn't.

The Compounding Asset: Data Nobody Else Can Generate

Why Benchmarks Grow More Valuable With Every Assessment

The fourth lever deserves its own section because it behaves unlike any other asset on your balance sheet. Equipment depreciates. Software gets cloned. A benchmarking database built from thousands of structured assessments does the opposite: it gains value with every new data point, at no additional investment, and it can't be copied without first copying the network that produced it.

The compounding is baked into the model. Each assessment adds a data point; each data point sharpens the benchmarks; sharper benchmarks make the next assessment more useful to the client who buys it. The asset appreciates as a side effect of the business operating.

When a buyer values a data asset, they're scoring it on four dimensions:

• Breadth. Coverage across industries and geographies. A dataset spanning 15 industry segments across 10 countries is worth exponentially more than one covering 3 segments in a single country.
• Depth. Volume per segment. Benchmarks become statistically meaningful at roughly 10 assessments per industry-geography intersection, and every assessment beyond that makes the comparisons more granular.
• Longitudinal reach. Time in the data. With three years of assessments you can show movement — say, automation maturity in financial services improving 18% year over year — which turns static benchmarks into predictive intelligence.
• Exclusivity. Your network generated it, through your methodology, inside your platform. A rival can't license it, scrape it, or synthesise it. The only path to an equivalent dataset is building an equivalent ecosystem.

For the ceiling on this lever, look at Gallup's CliftonStrengths: more than 30 million assessments, the largest dataset on human strengths in existence, and arguably the most defensible asset anywhere in the methodology business space — because matching it would require running 30 million assessments through a network you'd first have to build.

You will not get to 30 million. You don't need to. At 5,000 assessments across multiple industries, you hold something a buyer cannot manufacture — plus a trajectory toward 50,000 within a few years of acquisition. Acquirers price the trajectory of the data asset, not just its current size.

The Durability Question: Do People Stay?

The first four levers describe the value. The fifth tells a buyer whether that value will still be there in three years. Practitioner retention and client satisfaction are the two leading indicators that predict future revenue stability better than anything on the P&L.

Think of retention the way a property investor thinks about occupancy. If 92% of practitioners renew each year, the supply side of the platform is solid. If only 70% renew, the business sheds nearly a third of its capacity annually and burns constant energy on recruitment just to stay level.

How buyers read the retention number:

• Below 70%: a red flag. The ecosystem isn't delivering enough value to hold its own members, and the valuation gets discounted hard for churn risk.
• 70-85%: functional, not impressive. The network works but doesn't inspire loyalty; a buyer sees upside, and also sees the work required to capture it.
• 85-92%: strong. Practitioners stay because the network makes them more successful than independence would. This is the band where premium multiples live.
• Above 92%: exceptional. Switching costs are deeply embedded and the network sustains itself — practitioners are unlikely to leave even through a leadership transition, which is precisely the scenario a buyer is underwriting.

Client satisfaction is the companion signal, and the floor matters more than the ceiling. Track it per practitioner, per tier, and across the network. An average of 4.5 out of 5 with the lowest tier at 4.2 says every engagement meets a consistent standard regardless of who delivers it. Partner-tier practitioners at 4.8 while entry-tier sits at 3.2 says the brand promise depends on who picks up the engagement — a quality distribution problem that erodes everything else.

This lever comes last in the list but first in causality. High retention stabilises recurring revenue. High satisfaction drives referrals and feeds the data asset. Both together let the network effects compound. Let either slip, and every other lever degrades with it.

Five Levers, One Flywheel

The reason these levers are worth obsessing over is that they multiply rather than add. Retained practitioners generate recurring revenue; recurring engagements generate assessments; assessments deepen the data asset; the data asset strengthens the network effects; stronger network effects make the founder progressively less central — and a self-sustaining ecosystem retains practitioners better, which restarts the loop.

It also explains why the biggest business in a category isn't always the most valuable one. A EUR 1 million business with 90% recurring revenue, real network effects, proven founder independence, a growing data asset, and 90%+ practitioner retention can out-multiple a EUR 5 million business carrying 50% project revenue, a thin moat, and a founder it can't function without.

Which gives you a filter for every strategic decision from Day 1. Not "does this add revenue this quarter?" but "does this move at least one of the five levers in the right direction?"

Expect the timeline to feel uneven. Year 1 drags, because infrastructure is invisible work. Year 2 accelerates, because the flywheel catches. Year 3 starts to feel inevitable, because compounding has produced something genuinely hard to compete with. The founders who win at this play the long game: prove before scaling, build density before expanding, defend quality above everything — and then let the compounding do what compounding does.