The average multiple is not your multiple
There is a widely cited number for what middle-market companies sell for. It is real, it is well sourced, and it describes a set of transactions that probably does not include yours.
Somewhere in the first few weeks of thinking about a sale, most owners find a number. A multiple. Six point something, seven point something, times EBITDA. They multiply it against their own earnings, and from that moment forward they have a price in their head.
It is worth understanding where that number comes from, and who it describes.
Who GF Data is
GF Data is the most widely cited source for private-company transaction multiples in the middle market. It collects deal data from private equity firms and other deal sponsors — purchase price, EBITDA, leverage, terms — and publishes quarterly reports on what those transactions actually priced at. It is now part of ACG, the Association for Corporate Growth.
It matters because private company transaction data is otherwise almost impossible to get. Public companies disclose. Private ones do not. A seller trying to understand pricing is generally reading either public-company comparables, which are not comparable, or a broker's assertion, which is not data. GF Data is one of the few sources built from actual reported private transactions.
So when a number gets quoted, it is usually theirs, and it is usually being quoted for good reason.
The number
GF Data reported full-year 2025 purchase price multiples averaging 7.2x trailing twelve-month adjusted EBITDA. Within that year, the quarterly figures moved — Q2 came in near 6.9x, Q3 rose to 7.5x. Different quarters, different mixes of deals, different numbers.
That is the figure most people find.
Who that number describes
Read the definition of the dataset, which almost nobody does.
Those multiples cover private-equity-sponsored transactions with enterprise values between $10 million and $500 million.
Consider what sits at the top of that range. A $400 million transaction and a $12 million transaction are both inside the average. They do not price anything alike, and the larger one pulls the number.
GF Data reports the tiers separately, and the tiers are the useful part:
- Deals between $10 million and $25 million of enterprise value averaged around 5.9x through the third quarter of 2025. - Deals between $100 million and $250 million averaged around 10.0x. - Below $10 million of enterprise value, reported figures for the first half of 2025 ran closer to 5.5x.
Same dataset. Same period. A four-turn spread.
The size premium is the actual story
The gap between large and small is not noise, and it is not sentiment. GF Data's long-run spread between platform buyouts above $100 million of enterprise value and everything below it averages roughly 2.6 turns. Through the first nine months of 2025 it widened to about 2.8 turns — large platforms near 9.8x against roughly 7.0x for everything else.
There is a reason, and it is not that buyers like big companies better.
A company with $3 million of EBITDA and eleven customers carries a different risk profile than a company with $30 million of EBITDA and four hundred. If the largest customer leaves, the first business loses a third of its earnings and the second business loses a rounding error. Buyers price that difference. No growth story moves it.
The same logic runs through management depth, supplier concentration, geographic exposure, and the question of what happens when the owner stops showing up. Size correlates with all of it.
Two numbers that are not the same number
Before comparing any multiple to any other multiple, check what it is a multiple of.
EBITDA — earnings before interest, taxes, depreciation and amortization — assumes the business pays a manager to run it.
SDE — seller's discretionary earnings — adds the owner's full compensation and benefits back in, on the assumption that a buyer replaces the owner personally.
The same company, at the same price, can be honestly described as 3x SDE and 6x EBITDA. Both statements are true. Neither party is lying.
This is why the numbers a seller encounters seem irreconcilable. Main-street business brokers typically quote SDE multiples — BizBuySell reported an average cash flow multiple of 2.7x in the second quarter of 2026, against a median sale price around $349,000. Investment banks and private equity sponsors quote EBITDA multiples. An owner who reads both concludes that someone is being dishonest with him. Usually nobody is. They are measuring different things.
Ask which one any quoted multiple refers to. Ask every time.
What else moves it
Industry sets a band. Published lower-middle-market ranges for 2026 put home services somewhere around 4x to 6x, manufacturing around 5x to 7x, construction around 3x to 5x, distribution around 5x to 7x.
Then company quality decides where inside the band you land, and the spread inside a band is often as wide as the spread between bands.
Recurring revenue is the largest single lever in the trades. A service business with a substantial base of recurring maintenance contracts has been reported to trade a full turn or two above an otherwise similar business without one. That is the difference between an agreement a customer renews and a phone that rings when something breaks.
Customer concentration cuts the other direction, and hard.
Then there is everything neither you nor the buyer controls. Interest rates determine how much debt a buyer can service, which determines what they can pay. Private equity appetite for a sector runs in cycles — a trade that is being actively consolidated prices differently than one that is not, and that can change within a year. Multiples reported for 2024 and 2025 moved with credit conditions, not with the underlying businesses.
So what is it worth
The honest answer is that a published average tells you what a set of transactions you were not part of priced at, during a period that has already ended.
It is a starting point for a conversation. It is not a valuation, it is not a forecast, and it should not be the number you build a retirement around.
What it is genuinely useful for is direction. If you know the size premium exists, you understand why growing for two more years might be worth more than selling now. If you know recurring revenue moves the multiple, you know what to build before you go to market. If you know customer concentration cuts it, you know what to fix.
That is the value of the number. Not the number itself.
For the mechanics behind these figures — adjusted EBITDA, add-backs, and the factors that move a specific company inside a published range — see how buyers price a Texas trades business.
---
Sources: GF Data (an ACG company), quarterly and full-year 2025 middle-market M&A reports; BizBuySell Insight Report, Q2 2026; published lower-middle-market sector ranges as of 2026. Figures reflect data reported through mid-2026 and change with market conditions.
Last reviewed: August 2026.