Lower Middle Market EBITDA Multiples: Data by Deal Size and Industry

Anyone who starts looking at lower middle market companies runs into the same frustration early on: no two sources seem to agree on what these businesses are worth. One report puts a company at 3x earnings, another at 7x, an index says 10x, and after a while the whole idea of a “multiple” starts to feel arbitrary.

It isn’t. The multiple depends, more than on almost anything else, on how big the company is. In 2025, private equity buyers paid around 7.2x EBITDA for the average sponsored deal between $10 million and $500 million of enterprise value (GF Data), but that average hides an enormous range. A business selling around $20 million of enterprise value went for roughly 6.4x. One at $150 million cleared close to 10x. Below $10 million of enterprise value, you are down near 5.5x. Same market, same year, nearly four turns of spread, and size drives far more of it than industry does.

So the real question about any one of these companies is not whether its multiple sounds high or low. It is which size band the business sits in, and what about the business itself might move it a turn or two off the average once it is there.

This guide walks through lower middle market EBITDA multiples by size band and by sector, the mechanics that produce the size premium, and what moves an individual company’s number.

Lower middle market EBITDA multiples by deal size

The most consistent public series on lower middle market entry pricing comes from GF Data, which collects transaction detail from more than 330 North American private equity firms including funded and independent sponsors, family offices and mezzanine firms. Its core universe is private equity sponsored deals valued between $10 million and $500 million of total enterprise value, and all its multiples are stated as TEV over trailing twelve-month adjusted EBITDA.

Multiples by enterprise value

Two features of this table matter more than the headline number.

1. The gradient itself. A company sold at $20 million of enterprise value and a company sold at $150 million are not in the same pricing market, and treating “lower middle market” as one price point produces a number that describes neither.
2. The thinness at the top. The $250 million to $500 million row carries 104 observations across 23 years, which is why it swings from 10.9x to 8.5x between 2021 and 2025 in a way the deeper rows do not.

Multiples by company EBITDA

Enterprise value is a useful cut for a buyer. Sellers and operators usually think in earnings, and GF Data publishes that view separately.

A company with $5 million of adjusted EBITDA sat at 7.4x on average through the first three quarters of 2025, which implies an enterprise value near $37 million.

Below $10 million of enterprise value

GF Data reports separately on a $1 million to $25 million universe. Through the first half of 2025 it tracked 118 transactions there, with the $1 million to $5 million tier averaging roughly 5.5x and the $5 million to $10 million tier roughly 5.6x, against 6.2x to 6.7x for the $10 million to $25 million tier. Those figures come from a different universe and a different period than the tables above, so they sit beside those tables, not inside them.

In practice: close to a full turn of EBITDA separates sub-$10 million deals from the tier immediately above. GF Data measured that gap at 0.9 turns on buyouts through the first half of 2025, and at 5.5x against 6.4x through the first three quarters of 2024. The historical average spread is 0.7 turns.

In 2023 the relationship briefly inverted, with sub-$10 million deals pricing at 6.2x against 6.0x for the tier above.

What “EBITDA multiple” means in this market

The multiples above are total enterprise value divided by trailing twelve-month adjusted EBITDA. Enterprise value is the price for the business on a debt-free, cash-free basis, which is what makes multiples comparable across companies with different capital structures. Adjusted EBITDA is reported earnings after add-backs, and add-backs are where most of the negotiation happens.

The single most expensive error in this market is comparing an EBITDA multiple to an SDE multiple.

Why published multiples disagree

BizBuySell reported an average cash flow multiple of 2.7x in the second quarter of 2026, applied to seller’s discretionary earnings across main-street business-for-sale transactions, where the median sale price is roughly $349,000. The DealStats Value Index put the median selling price to EBITDA at 3.5x in the fourth quarter of 2025, covering private company transactions of all sizes including asset sales. GF Data’s sponsored-deal universe sat at 7.2x for the same year. Capstone Partners reported 9.8x, measuring middle market transactions with disclosed multiples.

Four providers, four defensible answers, spanning a factor of nearly four. None is wrong. They measure different companies, using different earnings definitions, over different periods, with different selection rules about which deals enter the sample.

The practical boundary sits around $2 million of transaction value. The IBBA and M&A Source define main street as $0 to $2 million and the lower middle market as $2 million to $50 million, and that line is roughly where the earnings basis flips from SDE to adjusted EBITDA, because it is roughly where a buyer stops assuming a working owner and starts budgeting a hired manager.

Add-backs and quality of earnings

Plenty of add-backs are legitimate. In our underwriting, two categories recur often enough that we look for them by default. The first is the replacement cost of the owner and any working family members. A seller will frequently price a single manager’s salary against an owner who is running sales, operations, and the customer relationships at once, and on the smaller companies that understatement can be material to the adjusted number.

The second is the one-time expense that is not one. A marketing campaign the seller judges to have failed is a favorite candidate for exclusion, and we do not accept the premise that attributing sales to specific tactics is too difficult to attempt.

Sell-side quality of earnings work has a measurable effect on price, though not uniformly. Across 360 transactions tracked from the third quarter of 2024, sellers who commissioned a sell-side QoE averaged 7.4x against 7.0x for those who did not, with the benefit concentrated above $50 million of enterprise value.

Why smaller companies trade at lower multiples

The size premium is the most durable pattern in this data. GF Data’s long-run spread between platform buyouts of $100 million to $500 million of enterprise value and those below $100 million averages 2.6 turns, and it widened to 2.8 turns through the first nine months of 2025, with large platforms at 9.8x against 7.0x for the rest. Four mechanisms produce it, and none of them is sentiment.

Concentration risk is arithmetic, not opinion

A company with $3 million of EBITDA and eleven customers has a different loss distribution than a company with $30 million of EBITDA and four hundred. The largest customer leaving costs the first company a third of its earnings and the second company a rounding error. Buyers price that distribution, and no amount of growth narrative moves it.

Management depth is a cost the buyer inherits

Smaller companies are more likely to depend on the departing owner for sales, pricing authority, supplier relationships, and institutional memory. The buyer either retains that person, replaces them, or absorbs the gap. Each of those has a cost, and each of them is uncertain at the point the price is set.

The buyer pool thins as companies get smaller

At $150 million of enterprise value a seller runs a formal auction against committed funds with capital they are obligated to deploy. At $15 million the pool is independent sponsors raising capital deal by deal, family offices, search funds, and individual buyers, several of whom face financing contingencies. Fewer credible bidders means less competitive tension, and competitive tension is a large part of what the top of the market is paying for.

Debt capacity does not scale linearly

Smaller deals support less leverage, and the equity that fills the gap is more expensive than the debt it replaces. The detail sits in the debt capacity section below, and it is the mechanism most directly visible in the transaction data.

GF Data’s contributors described the 2025 widening as a renewed preference for scale and creditworthy assets under tighter lending conditions. A size premium driven partly by credit is a size premium that moves when credit does.

Lower middle market multiples by sector

Sector moves the multiple by under two turns from the highest priced sector to the lowest, against close to four turns across the size bands. The table below gives the range each sector traded across from 2021 to 2024 alongside two separate 2025 cuts, because a single sector point estimate hides how much these numbers move.

Media and telecom reads 8.6x for nine months of 2025 and 6.9x for the full year. Nothing dramatic happened in the fourth quarter. The sector carries 85 observations across 23 years, so a handful of transactions moves the average almost two turns.

Sample depth explains most of what this table shows. Business services and distribution, with more than 1,300 and 600 observations respectively, hold within a range of a few tenths year after year. Media and telecom, retail and technology swing by two to three turns.

A sector average built on a thin sample describes the deals that happened to close. It does not price the next one.

Dispersion inside a sector also exceeds the gaps between sectors. GF Data’s sector-by-size breakout through the third quarter of 2025 put business services deals from $100 million to $250 million at 11.0x against 7.4x for the $25 million to $50 million band. Healthcare services ran 9.2x at $50 million to $100 million against 6.0x at $10 million to $25 million. A healthcare company at the bottom of the size range prices below a manufacturer at the top of it.

Platform and add-on multiples

The standard account of multiple arbitrage holds that a platform buys add-ons at a lower multiple than its own, and the spread accrues to the platform’s equity on exit. The data no longer supports the first half of that sentence.

Add-on share and add-on pricing

Inside GF Data’s sponsored $10 million to $500 million universe, add-ons ran at 40% of buyouts through the first half of 2025, down from a mid-2024 peak of 44%. Count-based measures of the whole US buyout market put the add-on share substantially higher, because they count a different denominator.

On pricing, add-ons briefly overtook platforms in the first quarter of 2025 at 7.7x against 7.6x, and by the first half of 2025 the two had converged at 7.2x each. GF Data has written that competitive conditions over the past two years drove add-on pricing up, in some cases above platform pricing.

Repeating the received wisdom will misprice a deal.

If a roll-up model assumes add-ons arrive a turn and a half cheap, and the market is clearing them level with platforms, the arbitrage line in that model is fiction. The measured record on roll-up returns is set out in the roll-up statistics report.

Where the add-on advantage actually sits

The platform and add-on gap is real in financing. In the first half of 2025, add-ons in the $1 million to $5 million tier carried debt coverage of 5.7x trailing EBITDA against 2.3x for standalone platforms of the same size. An add-on borrows against the combined entity’s cash flow and credit profile; a new platform at that size borrows against a single small company. That financing gap, more than any purchase price discount, is what makes small tuck-ins work for acquirers who already own scale.

The effect distorts other measurements. In the $10 million to $25 million band, sellers who ran a sell-side QoE averaged 5.9x, against 6.6x for those who did not. GF Data reads this as a mix effect: 63% of the non-QoE deals in that band were add-ons against 50% of the QoE deals, and add-on pricing swamped the process effect.

What moves an individual company’s multiple

Market averages set the starting point. The distance a specific company travels from that starting point is decided by a short list of characteristics, and in the current market the list has reordered itself.

Financial performance is barely being paid for

Companies with above-average financial performance received a 2% premium over other buyouts through the third quarter of 2025, 7.3x against 7.1x, the narrowest spread in GF Data’s tracked history. The high-performer premium, defined as revenue growth above 10% with EBITDA margins above 10%, had already compressed to roughly 5% in the first half of the year.

This is a real and uncomfortable finding for anyone arguing that quality earns its price. It says buyers in 2025 were paying for scale and creditworthiness ahead of growth and margin.

Revenue quality and customer concentration

Recurring or contracted revenue, low customer concentration, and demonstrable retention move a multiple more reliably than a growth rate does, because they change the distribution of outcomes instead of the midpoint. This is where a company most often justifies a price above its band.

Owner dependence and management continuity

A business that runs without its founder is worth more than the same business that does not, and the gap widens as the company gets smaller. Rollover equity is the market’s mechanism for handling this. Through the third quarter of 2025, 68.3% of completed platform deals included seller rollover, averaging 14.8% of total enterprise value. In the $1 million to $10 million tier through 2024, deals with rollover averaged 6.0x against 4.3x without it.

Process and preparation

Sell-side quality of earnings, clean financials, and a competitive process are worth roughly half a turn above $50 million of enterprise value, and close to nothing below it.

What this means: below $50 million of enterprise value, preparation buys certainty of closing. It does not buy a higher headline price. That is still worth doing, and it is not worth budgeting a multiple premium for.

A structured approach to assessing these factors on a specific transaction is set out in the guide to evaluating an independent sponsor deal.

Debt capacity, rates, and what they do to pricing

Leverage is the mechanism that connects the credit market to the price a seller receives. When lenders advance less against the same earnings, the buyer either writes a larger equity check or bids less.

Leverage by size band

The aggregate total debt figure rose by a tenth of a turn between 2024 and the first three quarters of 2025, and four of the five size tiers declined over the same period. Averages built on shifting mix behave like that, which is why the rows carry the signal here. The full year settled the point: total debt across the same universe closed 2025 at 3.6x, below 2024’s 3.7x (CIBC US Middle Market Monitor, Q1 2026, GF Data-sourced).

Equity has carried roughly half of total capitalization every year since 2021, moving between 49.0% and 52.1%, with senior debt between 38.8% and 42.5% and subordinated debt between 8.0% and 9.1%. That stability is easy to miss in commentary about tightening credit. What moved was not the proportion of debt so much as its price.

The cost of debt

Average senior debt pricing on deals from $10 million to $250 million of enterprise value ran 10.4% at the end of 2023, fell to 8.1% by the first quarter of 2025, and returned to 8.6% by the third quarter. Ninety-day trailing SOFR sat at 4.3% as of GF Data’s February 2026 commentary, with first-lien leveraged loan spreads typically running 550 to 700 basis points over it.

What was once a 6% cost of debt now approaches 10%.

Direct lenders and business development companies have taken share as banks remain constrained by regulatory capital requirements, and unitranche structures that previously carried light covenants now include maintenance tests and amortization.

What the lender type changes

Pricing also differs by lender type in a way that matters at this size. On platform deals from $10 million to $50 million of enterprise value, bank senior debt priced at 7.5% against 12.0% for unitranche. Those cells sit inside samples of 41 bank and 49 unitranche platform observations spread across two size tiers, so treat the gap as directional.

The unitranche premium buys certainty, speed and a single counterparty. It also removes roughly four and a half points of return from the capital structure before the equity sees anything.

What the entry multiple does and does not tell you about returns

A lower entry multiple is a real advantage and it is not the same thing as a better investment. The arithmetic shows why.

A worked comparison at two entry multiples

Take two companies with identical $4 million of adjusted EBITDA. Company A is bought at 5.5x for $22 million, Company B at 6.5x for $26 million. Both borrow 3.5x EBITDA, so both carry $14 million of debt at close, leaving $8 million of equity in A and $12 million in B. Assume both pay debt down to $7 million over the hold and both exit at 7.0x.

If both grow EBITDA to $6 million, both exit at $42 million of enterprise value and $35 million of equity value. A returns 4.4x its equity. B returns 2.9x. The turn of entry multiple cost B a third of the outcome.

Now change one variable. If A grows to $5 million while B grows to $8 million, A exits at $35 million of enterprise value for $28 million of equity, a 3.5x return. B exits at $56 million for $49 million of equity, a 4.1x return. The company that cost more at entry produced the better result, and it did so without any assumption about multiple expansion.

Illustrative arithmetic only. Figures are hypothetical and exclude fees, taxes, transaction costs and interim cash flow. The 7.0x exit assumption sits above the 6.8x average for the $25 million to $50 million band through Q3 2025, and Company B’s exit at $56 million would land it in the $50 million to $100 million band, which averaged 8.3x over the same period. Growing into a higher size band is itself part of how these returns are made, and it is not guaranteed.

Why we do not lead with the multiple

We think the raw multiple is the wrong first question to ask about one of these companies. A business bought at a more expensive multiple can be the better deal outright if its growth rate is vastly superior, or if the quality and dependability of its revenue carry far more visibility. If the job were simply finding the cheapest multiple in the market, the job would be easy. It is not.

That view sits in genuine tension with the market data above, and the tension is worth naming instead of smoothing over. Buyers through 2025 paid a 2% premium for above-average financial performance. An underwriting approach that pays up for growth and revenue quality is, on those numbers, buying something the market is currently discounting. Whether that reads as an opportunity or a warning depends on how long the capital is staying invested and whether the buyer is right about the company.

A lower entry multiple is a real advantage and it is not the same thing as a better investment.

Entry price is one of three levers, alongside earnings growth and exit multiple, and it is the only one fixed on day one. That is why it is worth arguing about, and why it cannot carry an underwriting case on its own. The risks specific to this segment, including the dispersion that makes any single entry multiple a weak predictor, are treated at length in the lower middle market risks guide. Hold periods, which determine how long the arithmetic above has to work, are covered in the holding period statistics report.

How CapitalPad invests in lower middle market private equity

CapitalPad is a private equity co-investment group that gives accredited investors a way to invest in lower middle market private equity one deal at a time. Rather than committing capital to a fund and a blind pool, an investor sees each opportunity on its own terms, a specific company at a specific price with a sponsor already behind it, and decides whether to invest or pass.

The businesses sit at the bottom of the size range this guide covers: established, historically profitable US and Canadian companies with $1 million to $7 million of EBITDA and $5 million to $30 million of enterprise value, the tier where entry multiples run lowest. CapitalPad underwrites each transaction before it reaches the deal room, pools the investors in a given deal into a single deal-specific SPV, and sets minimums from $25,000, so an investor can build lower middle market exposure one position at a time instead of through a single large fund commitment. Because there is no fund, there are no capital calls and no deployment pressure behind any deal shown, and full investor economics are published on CapitalPad’s FAQ.

The population, leverage, and return data behind the segment is collected in the lower middle market statistics report, and the asset class itself is laid out in the lower middle market private equity guide.

FAQ

What multiple would a company with $2 million of EBITDA sell for?

That sits below the floor of most sponsored-deal datasets, so answering it means stitching two sources together. At 5x to 6x, a company with $2 million of EBITDA is a $10 million to $12 million transaction, which sits at the very bottom of GF Data’s tracked universe.

Its $1 million to $25 million series showed sub-$10 million enterprise values averaging roughly 5.5x in the first half of 2025 and the $10 million to $25 million tier at 6.2x to 6.7x. Expect the low-to-mid 5s if the business still depends on its owner, and the mid 6s if it has a management layer, contracted revenue and clean financials.

Why is a broker quoting a higher multiple than these tables show?

Three things are usually happening at once, and they compound.

1. Asking prices are not clearing prices.
2. Broker quotes below roughly $2 million of value are frequently stated on seller’s discretionary earnings instead of EBITDA, which mechanically produces a lower-looking multiple on a higher-looking earnings figure.
3. Indices built from deals with publicly disclosed multiples, such as Capstone Partners’ 9.8x average for 2025, tend to skew upward, since disclosure correlates with size and deal quality.

Ask which earnings figure the multiple is applied to before comparing it to anything.

Sources

1. GF Data, “ESOP Advisor Special Report, Q3 2025.” Charts 1 to 4 and 9; private equity sponsored deals $10M to $500M TEV (industry series $10M to $250M TEV); average TEV/TTM adjusted EBITDA, debt multiples, and industry multiples. https://gfdata.com/wp-content/uploads/Q3-25_GFData_ESOP_Report.pdf
2. GF Data, “Small-Deal Resilience: Why the Under $25 Million Tier Still Moves in H1 2025” (2025). Sponsored transactions, $1M to $25M TEV, H1 2025; small-tier multiples and add-on debt coverage. https://gfdata.com/small-deal-resilience-h1-2025/
3. GF Data, “The Size Premium Returns to 2.8x” (2026). Large-versus-small platform spread and the above-average financial performance premium, from GF Data’s Q3 2025 M&A Report. https://gfdata.com/size-premium-esop-competitive-advantage/
4. Middle Market Growth (Association for Corporate Growth), “New Fields, New Fruit: Digging Into GF Data’s Newly-Added Data” (October 2025). Sell-side quality-of-earnings pricing across 360 transactions, Q3 2024 to Q2 2025. https://middlemarketgrowth.org/fall-2025-gf-data-newly-added/
5. GulfStar Group, “Middle Market Commentary, 1H 2025” (September 2025). GF Data Q2 2025 M&A Report, cited for add-on share of buyouts and add-on-versus-platform pricing. https://gulfstargroup.com/middle-market-commentary-first-half-2025/
6. CIBC US Middle Market Monitor, Q1 2026. GF Data-sourced; sponsored LBOs $10M to $500M TEV; full-year 2025 entry multiples, sector multiples, and leverage.
7. BizBuySell Insight Report, Q2 2026. Main-street business-for-sale transactions; average cash flow multiple on seller’s discretionary earnings, and median sale price. https://www.bizbuysell.com/insight-report/
8. DealStats Value Index, Q4 2025 (Business Valuation Resources). Private company transactions of all sizes including asset sales; median selling price to EBITDA. https://www.bvresources.com/articles/bvwire/dealstats-value-index-ebitda-multiples-ease-in-late-2025
9. Capstone Partners, “2025 Middle Market M&A Valuations Index” (April 2026). Middle market transactions with disclosed multiples; average EV/EBITDA. https://www.capstonepartners.com/insights/report-capstone-partners-middle-market-mergers-and-acquisitions-valuations-index/
10. IBBA and M&A Source, Market Pulse Survey. Market-segment definitions: main street $0 to $2M, lower middle market $2M to $50M. https://www.prnewswire.com/news-releases/the-ibba-and-ma-source-announce-the-market-pulse-q4-2025-survey-results-302691992.html
11. CapitalPad, Official Information About CapitalPad. https://capitalpad.com/official-information-about-capitalpad/