To value an acquisition target, rebuild its earnings yourself, strip the seller's adjusted EBITDA down to what survives diligence, then apply a multiple that reflects measurable quality: growth durability, customer concentration, owner dependence and margin versus the category norm. In the lower mid-market that typically means 3.5x-9x adjusted EBITDA for consumer and services businesses, 6x-15x for software, and roughly 1.8x-4.2x SDE once earnings fall below about $500k. Most buyers overpay not because they choose the wrong multiple, but because they apply a defensible multiple to an indefensible earnings number. The method below fixes the number first, then the multiple, then the structure.
Why can't you take the CIM's EBITDA number?
Because a CIM is a sales document, not an accounting record. Adjusted EBITDA in a CIM is operating profit restated to show what the business earns for a new owner, and every add-back the seller persuades you to accept is worth 3.5x to 9x its value in price. That incentive reliably produces "one-off" marketing spend that recurs every year, "non-recurring" legal fees that appear in all three years of accounts, and an owner salary replaced with a number no competent general manager would accept.
In our deals, the gap between the CIM number and the number that survives diligence is routinely 10-25% of stated EBITDA. On a business marketed at £2m EBITDA and a 6x ask, that gap is £1.2m-£3m of price. Rebuild the number from source, management accounts, bank statements, payroll and tax filings, never from the deck.
How do you normalise the earnings yourself?
Work through four moves, in order:
- Start from statutory profit, not the seller's bridge. Reconcile it to actual bank movements over the trailing twelve months before you adjust anything.
- Accept only provable non-recurring costs. The test is the same one we apply on the sell side: would this cost genuinely not exist under new ownership, and can the seller prove it? A settled, documented litigation passes. "Exceptional" ad spend in a business that must advertise to grow does not.
- Replace owner economics with market rates. Add back the founder's £300k salary, then deduct the £120k it costs to hire the manager who replaces them. Buyers routinely do the first half and forget the second.
- Deduct what new ownership adds. Market rent if the seller owns the premises personally, software the business has been gifted, family members working below market or for free.
For the anatomy of which adjustments hold and which collapse under scrutiny, see our guide to adjusted EBITDA add-backs, written for sellers, which is exactly why it is useful to buyers.
Which quality factors should move the multiple?
Within any category band, four factors decide whether you pay the bottom or the top of the range:
- Growth durability. Three years of compounding organic growth earns the top of the band. Growth bought with escalating paid acquisition is rented revenue, and prices one to two turns lower, you are inheriting the rent.
- Customer concentration. A top customer above 20% of revenue costs at least a turn. Above 40%, treat the concentrated revenue as contingent: structure it, don't pay cash at completion for it.
- Owner dependence. If the founder holds the key relationships, the technical knowledge or the public face of the brand, part of what you are buying leaves at completion. A business that demonstrably runs on its second-tier team commands a premium of a turn or more.
- Margin versus category norm. Margins far above the category norm often signal underinvestment you will have to fund. Margins below it can be genuine upside, but you pay for the business you diligence, not the one you plan to build.
These overlap heavily with the red flags we screen for when buying a business. The difference is that a red flag kills the deal; a quality discount reprices it.
What do lower-mid-market businesses actually sell for?
Typical lower-mid-market ranges, applied to properly normalised earnings:
- Consumer brands. 3.5x-9x adjusted EBITDA by category and size; repeat-purchase categories with genuine subscription revenue sit at the top.
- Services businesses. 3.5x-9x; contracted, recurring revenue trades several turns above project work won deal by deal.
- Software and apps. 6x-15x EBITDA, or 2x-4.5x revenue where growth and retention justify pricing on the top line.
- Agencies. 3.5x-7.5x, with retained clients and low founder-sold revenue at the upper end.
- Distribution. 3.5x-7x, driven by exclusivity of supplier relationships and switching costs.
- Sub-$500k earnings. Below roughly $500k of earnings, buyers price on SDE at around 1.8x-4.2x, because owner economics dominate the P&L.
How do you sanity-check the price against comparables?
A multiple defended only by the seller's advisor is an opinion. Test yours against transactions in the same category and the same size band, size matters as much as sector, because a £10m-revenue business and a £50m-revenue business in the same category trade turns apart. Our comparable transactions tool is built for exactly this check, and our active deals show live terms in the lower mid-market. If your price sits above the comparable range, you need a written reason, a synergy you control, not one you hope for.
How should you structure around residual risk?
After diligence there will still be risk you cannot price away: the big customer might leave, the founder's relationships might not transfer, the growth might slow. The mistake is paying cash today for certainty you don't have. Structure it instead:
- Earnout. Make 10-30% of headline value contingent on revenue or gross-profit targets over one to three years. Tie it to metrics the seller can still influence post-completion, or you are inviting a dispute.
- Holdback or escrow. Retain 5-15% for 12-24 months against warranty claims and working-capital true-ups.
- Seller note. Defer 10-30% at a modest coupon. It bridges valuation gaps and keeps the seller financially invested in a clean handover.
In our deals, structure is where price gaps close. A seller convinced their business is worth 7x and a buyer convinced it is worth 5.5x can both be right, the difference is who carries the risk of the next two years.
What to do next
Run your target's numbers through our free indicative valuation tool, the same 10-question model we use on the sell side, and see how the quality factors above move the price. If you want vetted lower-mid-market deal flow rather than picked-over listings, our buyer network gives you access to founder-led businesses before they go wide.
What is your business actually worth?
Run your numbers through our free valuation tool. Ten questions, an indicative range built on real transaction multiples, and the exact build-up behind it.
Get your free valuation ↗Frequently asked questions
How do I choose the multiple to offer for a target?
Start from the sector band, 3.5x-9x adjusted EBITDA covers most lower-mid-market consumer and services deals, then place the target within it using measurable quality: growth durability, concentration, owner dependence and margin versus category norm. Price the business you will own after the founder leaves, not the one in the CIM.
How do you calculate the value of a company you want to buy?
Rebuild adjusted EBITDA from statutory accounts and bank data rather than the seller's CIM, accepting only add-backs that provably disappear under new ownership. Apply a category multiple, typically 3.5x-9x in the lower mid-market, adjusted down for concentration, owner dependence and paid-for growth. Then sanity-check against comparable transactions in the same size band and structure residual risk with an earnout or holdback.
Should I use EBITDA or SDE to value an acquisition?
Use SDE when earnings are below roughly $500k, because the owner's own labour dominates the profit; buyers typically pay 1.8x-4.2x SDE at that size. Above that, use adjusted EBITDA with a full market-rate management cost deducted, that is what lenders and later acquirers will price on, at typical lower-mid-market multiples of 3.5x-9x depending on sector.
When should a buyer push price into an earnout?
When the valuation gap is a belief gap: the seller’s number depends on growth, retention or margin that has not happened yet. Move 10-30% of price into an earnout tied to the specific metric in dispute, over one to three years. If the risk is fundamental rather than unproven, cut the headline price instead, earnouts compensate uncertainty, not weakness.
How does customer concentration affect what I should pay for a business?
A top customer above 20% of revenue usually costs at least one turn of EBITDA. Above 40%, treat the concentrated revenue as contingent, pay for it through an earnout or holdback rather than cash at completion. The question is not whether the customer is happy today, but whether the contract, relationship and pricing survive a change of ownership.