Lower-middle-market multiples: the published data

What do the published multiples show by deal size?

By deal size, the published multiples climb from 2.0 times SDE for businesses sold under $500,000 to 5.8 times EBITDA at $5 million to $50 million (IBBA and M&A Source, Q2 2026), and to 9.7 times EBITDA for private equity deals of $250 million to $500 million (GF Data, 2024 figures published in 2025). Each dataset that splits by size shows the same staircase, but on its own measure, so a row is read against the rows of its own source and not against the others.

Business Valuation Resources' DealStats database sizes companies by revenue rather than price, and reports medians: private companies with net sales under $1 million sold for a median 3.0 times EBITDA in 2024, and those over $10 million for 7.0 times (DealStats Value Index, Q1 2025). GF Data, which counts only private equity deals, reported averages of 6.4 times for its smallest tier and 9.7 times for its largest in 2024.

Across a year, GF Data's private equity buyers paid an average 7.2 times adjusted EBITDA for deals of $10 million to $500 million in 2025, the same as in 2024 (ACG, February 2026). Capstone Partners' index of US middle-market deals with a disclosed multiple averaged 9.8 times in 2025 (Capstone Partners, 2026).

Why does the measure change from SDE to EBITDA at about $2 million?

Because the two sizes are priced on different earnings: IBBA and M&A Source report deals under $2 million of purchase price as a multiple of seller's discretionary earnings (SDE), and deals from $2 million to $50 million as a multiple of EBITDA (Market Pulse, Q2 2026).

SDE is the profit with one owner's pay and personal costs added back, which is how a buyer who will run the business reads it, so it is the larger number for the same business. EBITDA counts a market salary for the role the owner fills, which is how a buyer who will hire someone to fill it reads the business. That is why 3.1 times SDE at $1 million to $2 million and 4.0 times EBITDA at $2 million to $5 million are not a step of 0.9 on one scale: they are taken on two different numbers.

Which number a price is built on is decided by the business and its buyers, not by the seller's choice. The buyer of a $10 million company will rebuild EBITDA from the books in diligence, and the adjusted figure it accepts is the one the multiple is applied to.

Why do the published multiples disagree?

They disagree because each dataset counts different deals in a different way: who reports them, which buyers they cover, whether the figure is a median or an average, and whether size means the price, the enterprise value, or the revenue.

• A median and an average are not the same figure. A few large sales pull an average up, so an average multiple usually sits above the median for the same deals.
• Private equity is one kind of buyer. GF Data counts only private equity deals, so it says nothing about what a competitor or a family office paid.
• A disclosed multiple is a narrower set. Capstone's index counts only deals whose multiple was disclosed, which is a smaller set than every deal that closed.
• Revenue is not price. A company with $10 million of net sales in DealStats may sell for far less than the $10 million deal at the bottom of GF Data's range.

No one dataset is the right one. Each answers a narrower question than its headline suggests, and the reading the three that split by size share is the direction: the multiple rises with size, and the spread at any one size is wide.

What did buyers pay more for, beyond size?

In GF Data's private equity deals, buyers paid more for better financials: buyouts of companies GF Data classes as having above-average financials averaged 7.7 times EBITDA in 2024 through the third quarter, against 6.6 times for the rest (Private Equity Mid-Market Overview, 2024).

A sell-side quality of earnings report went with a higher price only in the larger deals. Across 360 private-equity-backed deals from Q3 2024 to Q2 2025, those with one averaged 7.4 times EBITDA against 7.0 times without; below $50 million the order reversed, 5.9 times with one against 6.6 times without at $10 million to $25 million (GF Data, 2025).

Both are comparisons of the deals that happened, not proof that one thing caused the other. What they show is what the buyers in that data paid more for: better financials, and, in the larger deals, earnings tested before the buyer arrived.

What should a founder take from the published multiples?

That a published multiple is the middle of a wide range, and where one company lands in it is decided by evidence a buyer can verify. Two chapters of the FNDRS Playbook are built on that: the multiple range (F02), where a sector's band is public and the founder's place in it is not, and the size premium staircase (F09), where a larger business earns a higher multiple for what its size proves to lenders and buyers.

None of the figures on this page is a valuation of any business, and FNDRS does not turn them into one. What a buyer would pay for yours depends on its earnings as a buyer rebuilds them, the risks it finds, and how many buyers compete for it. The Go-To-Market Analysis (GTMA) builds that file on your business before a buyer does.

• The multiple range (F02)
• The size premium staircase (F09)
• The Go-To-Market Analysis