GF Data, a survey of private equity-backed middle market deals, reported an average all-industry EBITDA multiple of 7.0x for the second quarter of 2026, down from 7.3x in the first quarter. Capstone Partners, using a different dataset, reported a full-year 2025 average of 9.8x. Both numbers are accurate, and an owner who picks the wrong one as a reference point can misjudge what a business is likely to sell for by millions of dollars.
EBITDA stands for earnings before interest, taxes, depreciation, and amortization, a common measure of a company's operating profit. A multiple expresses the purchase price as a number of years of that profit: a company with $5 million of EBITDA that sells for $35 million sold at 7.0x. Published multiples are averages of reported deals, so they depend entirely on which deals the publisher can see.
GF Data collects transactions from private equity groups and lenders, and Mercer Capital's fall update, which relays the figures, shows the all-industry average moving from 6.9x in the fourth quarter of 2025 to 7.3x in the first quarter of 2026 and back to 7.0x in the second, on roughly 80 to 85 deals per quarter. Capstone's index tracks annual averages across a broader set of reported transactions, and put the figure at 9.0x for 2023, 9.4x for 2024, and 9.8x for 2025. Capstone also reported that deals priced in the low double digits made up 40.7% of disclosed multiples in 2025, up from 31.6% in 2024. A dataset with more large, high-quality companies will report a higher average than one with more mid-sized and smaller ones.
The quarter-to-quarter swing in the GF Data figures is modest. A 0.3x decline from 7.3x to 7.0x on a company with $5 million of EBITDA is worth about $1.5 million in price, which is meaningful to an owner but well within the range of normal variation for a sample of 85 deals. Mercer described deal volume and average pricing as relatively stable compared with recent quarters, and the same description fits Capstone's long upward drift over three years.
The more useful signal is dispersion. Capstone noted that buyers accelerated diligence on lower-quality assets in 2025 while competition stayed strong at the upper end of the market. When buyers sort companies this sharply, the average falls between two very different outcomes, and few real businesses sell at the average. A company with recurring revenue, diversified customers, and a management team that can operate without the founder may price near the high end of the range. A company that depends on one customer or on the owner personally may price well below the average, however strong its EBITDA looks.
Expectations for 2026 reflect the same split. In a survey by Capstone and IMAP, advisors expected a typical business to trade at about 6.8x and a premium business at about 9.8x, with 66% expecting little or no change in multiples this year and 27.4% expecting an increase. The 3.0x gap between typical and premium in that survey is larger than the quarter-to-quarter movement in any published average.

Start with size. Multiples rise with scale because larger companies are more diversified, carry more management depth, and attract more buyers. A business with $2 million of EBITDA belongs in a different comparison group than one with $20 million, and a published all-industry average blends them together. When a survey reports multiples by EBITDA band or by industry, use that cut rather than the headline number.
Next, consider the buyer. Private equity groups, strategic buyers (operating companies acquiring for their own business), and individual buyers financing a purchase with a bank loan apply different return requirements and pay different prices for the same earnings. A strategic buyer that can combine the company with its own operations may pay more because it can remove costs, while an individual buyer is limited by what the business can support in loan payments.
Finally, test the earnings the multiple is applied to. A multiple is only as reliable as the EBITDA beneath it, and buyers typically adjust reported profit through a quality of earnings review, which tests whether the profit is recurring and supported by cash. Owner compensation, one-time expenses, related-party arrangements, and customer concentration are the usual adjustment areas. If a buyer's adjusted EBITDA is lower than the owner's, the same multiple produces a lower price.
Two developments will shape the third-quarter data. The first is financing: lender terms on acquisition loans affect how much buyers can pay, and tighter credit tends to pull multiples down for smaller companies first. The second is the mix of buyers. When private equity activity slows, the surveys built from private equity transactions can show lower averages simply because fewer large deals are in the sample, even if prices for comparable businesses have not changed. Owners should read each new release with that composition question in mind.
Published multiples are averages of different groups of deals, which is why a 7.0x and a 9.8x can both be correct. The benchmark that matters is the one built from companies like yours, applied to earnings a buyer will accept. Before using any survey number to set a price expectation, identify your size band, your likely buyer, and your adjusted EBITDA, and expect the result to land closer to the typical range than the premium range unless your business shows premium characteristics.