What the Median MSP Deal Actually Looks Like (2026)

The median MSP acquisition in 2026 is a business doing $3M to $8M in revenue selling to a regional buyer for somewhere between 5 and 9 times EBITDA, most of it cash at close, a chunk carried in a seller note, maybe an earnout tied to client retention, and a real chance the whole thing dies at the letter of intent before anyone signs a purchase agreement. It is a bolt-on. It closes quietly, and there is no press release, so you never read about it.

What you read about instead is the platform recap: Evergreen's 47th add-on, a $400M revenue provider getting recapitalised, a sponsor selling to a bigger sponsor at a double-digit multiple. That deal is real and it matters, but it is the exception, not the pattern. On M&A Signal's tracked sample, bolt-on acquisitions are 67 percent of activity and platform recaps are 29 percent. The coverage runs the other way around, because the platform deal comes with a banker, a comms team, and a number worth printing, while the bolt-on comes with an owner who signed an NDA and wants it kept quiet.

I read this from an operator's seat, because I run growth for MSPs and spend my time on where the money in this market actually moves. This piece walks the median deal end to end, the way I would explain it to an owner sitting across the table: what it is worth, how it is structured, how long it takes, and where it falls apart. One honesty note before any number: no party publishes an audited registry of private MSP transactions. Every stat here, including the ones from the most-cited sources, is assembled from disclosed deals, advisory-firm pipelines, and industry commentary. Treat the ranges as directional, not precise.

The deal that actually happens

Start with how many deals there even are, because the sources cannot agree, and the disagreement is the first thing worth understanding. Two of the most careful trackers count 2025 and give two different numbers.

M&A Signal counts 466 total MSP deals closed across North America in 2025, with more than $4.3B in disclosed value and volume up about 20 percent year over year. That is the broad industry aggregate. CT Acquisitions, in its "Private Equity MSP 2026" guide, counts 169 disclosed MSP M&A transactions in the same year, 69 percent of them involving private equity. That is the disclosed-only, PE-weighted slice.

These are not contradictions. They are two different scopes: a wide aggregate and a disclosed-deals subset. The mistake is picking one and calling it "the" 2025 total. Nobody has the clean number, because there is no registry to pull it from. What both agree on is direction: volume is high and rising, and the overwhelming majority of it is small.

On M&A Signal's own breakdown of 63 tracked transactions, 42 were bolt-on acquisitions (67 percent), 18 were platform PE recapitalisations (29 percent), and 3 were strategic mergers (5 percent). One caveat that is actually part of the argument: that 63-deal sample, by the firm's own methodology note, intentionally over-represents the deals that set pricing benchmarks. So even the best-known MSP M&A dataset skews toward the deals that get covered, and it still comes out two-thirds bolt-on. The real market, the one that includes every quiet sub-$5M sale nobody tracks, is almost certainly more bolt-on-heavy than that, not less.

A bolt-on is the base of the roll-up pyramid. A platform buys a smaller MSP in its region, folds the clients into an existing operation, strips duplicate overhead, and moves on. Evergreen's internal cutoff is the cleanest operational definition I have seen: operators above roughly $1M of EBITDA get treated as platforms, and everyone below that is an add-on. Most owners selling this year are on the add-on side of that line whether they think of themselves that way or not. The buyer taxonomy matters here, because the regional consolidator or search fund buying a bolt-on prices very differently from the institutional platform chasing a competitive process.

What the median deal is worth

Here is where the honesty line earns its keep, because the multiple you will actually get is the most-searched and least-settled number in this whole market. Four credible sources publish MSP valuation ranges, and for the sub-$10M band they disagree by two to three full turns of EBITDA. That spread is not sloppiness. It is what happens when every dataset has a different sample and no one can audit anyone else's.

M&A Signal's deal-size table puts a $2M to $5M ARR MSP at 5 to 7 times EBITDA ("small bolt-on, adds client density"), a $5M to $15M business at 7 to 9 times ("core acquisition target, the sweet spot for competitive processes"), and only the $40M-plus platforms at 9 to 13 times. Its market-wide median across all sizes is 9.0x, which tells you how much the big deals pull the average up and away from what a sub-$10M owner will see.

CT Acquisitions publishes two tables that do not even use the same earnings basis. Its "MSP M&A Multiples Report 2026" runs on SDE at the low end: sub-$1M revenue at 3.0 to 7.0 times SDE, $1M to $3M at 4.0 to 8.0 times, and $3M to $10M at 6.0 to 11.0 times adjusted EBITDA. Its separate "Private Equity MSP 2026" guide runs on EBITDA and lands lower for the same small businesses: under $5M EBITDA at 3.5 to 5 times, $5M to $15M EBITDA at 4.5 to 8 times, and only $15M-plus reaching 11 to 15 times and up. You cannot merge an SDE-based number and an EBITDA-based number into one figure. SDE adds back the owner's salary; EBITDA does not. Quote across those bands without flagging the basis and you will mislead someone by a full turn.

N2M Capital Advisors, a fourth and independent dataset of 120 analysed transactions, reports a median EV/EBITDA of 8.9x with a range from 4x at the sub-$5M-revenue end up to 14x for scaled platforms with strong recurring revenue and a cybersecurity story. Every one of these is a single-source, non-audited number. That they all disagree is the point: it reinforces, rather than undercuts, the fact that nobody owns audited MSP deal data. All advisory-firm multiples here are synthesis estimates blended from disclosed deals and pipeline observation, not audited transaction records.

My operator read on that table: the honest planning range for a clean, sub-$10M MSP with decent recurring revenue is a mid-single-digit EBITDA multiple, call it 5 to 8 times, and you should assume the low end until a competitive process proves otherwise. The double-digit multiples in the headlines belong to $15M-plus platforms with 80-percent-plus MRR share and an integrated cybersecurity practice, which most sellers are not. This is the exact inversion of what platform-deal coverage implies: the most common deal sits in the lowest-multiple band, not the highest. If you want the mechanics behind any single figure, the MSP valuation multiples breakdown and the how to value an MSP methodology walk the inputs.

How the median deal is structured

The multiple is the headline. The structure is where the number you actually receive gets decided, and for a bolt-on it is rarely all cash. The standard shape at the small end is cash at close, plus an escrow holdback, plus a seller note, plus an earnout, and each piece moves real money out of your day-one proceeds and into a later, conditional bucket.

On CT Acquisitions' structure breakdown, cash at close for sub-$5M SDE targets runs 70 to 90 percent of enterprise value, versus 85 to 95 percent for platform-tier deals. So the smaller you are, the more of your price gets deferred. Seller notes at the small end run a 3 to 5 year term at a 5 to 8 percent coupon, sized at 5 to 15 percent of enterprise value, and subordinated to the buyer's senior debt, which means if the deal goes sideways you are behind the bank in line. MRR-retention earnouts get measured over 12 to 24 months against client-cohort retention, sized at 5 to 20 percent of enterprise value, with the heaviest usage at sub-$3M where the buyer sees the most owner-dependency risk. Rollover equity, if it is in the deal, runs 5 to 15 percent on an add-on. Escrow is a standard 5 to 10 percent of enterprise value held for 12 to 18 months, and representations-and-warranties insurance starts showing up above a $3M enterprise value.

The earnout is the piece most sellers misjudge, so price it with your eyes open. SRS Acquiom's deal-terms data, which is broad private-target M&A rather than MSP-specific and should be read as the best available proxy, finds that earnouts pay out at roughly 21 cents on the dollar of their theoretical maximum on average. Roughly a quarter to a third of earnouts pay nothing at all, and even among the ones that pay something, only about half the maximum dollars come through. About 28 percent of earnouts end up contested. General MSP deal-structure practice ties the trigger to net-revenue retention, where holding 90-percent-plus of the acquired client cohort over 12 months pays the full earnout and dropping below roughly 80 percent zeros it out. So when a buyer offers you a headline number with 20 percent of it in an earnout, the honest expected value of that slice is closer to a few cents on the dollar than the full figure, and you should negotiate as if the earnout may never arrive.

Where the deal dies, and how long it takes

Before any of that structure matters, the deal has to survive to close, and a meaningful share of them do not. CT Acquisitions puts the failure rate at roughly 1 in 3 signed LOIs failing to close, explicitly labelled a directional industry estimate rather than an audited figure, because no public LOI registry exists. That is the number to plan around: sign an LOI and there is a real, roughly one-third chance you are back to square one a few months later, having spent legal and advisory fees and taken your eye off the business during diligence.

Where do they die? The best available breakdown comes from the Axial Dead Deal Report 2025, a study of 75 unsuccessful deals, and the important caveat is that it is lower-middle-market broadly, not MSP-specific. CT applies it to the MSP context, but the underlying sample includes non-MSP businesses, so read it as shape rather than precision. On that data, the leading killers are non-QoE diligence findings at 25.3 percent, quality-of-earnings EBITDA discrepancies at 21.3 percent (up sharply from 10.6 percent in 2023), retrade or renegotiation at 14.7 percent, seller withdrawal at 13.3 percent, financing constraints at 10.7 percent (down from 21.3 percent as credit eased), and business underperformance surfacing mid-diligence at 8.0 percent.

Two of those are worth an MSP owner's specific attention. The first is the QoE EBITDA discrepancy, which more than doubled as a cause of dead deals in two years. In plain terms: the buyer's accountants recast your earnings and the number comes back lower than the one you built the price on, and the deal either retrades down or dies. The single most disputed MSP-specific item, per the same report, is how you recognise deferred revenue on annual prepaid contracts. Get your revenue recognition clean and defensible before you go to market, or a diligence team will find the gap and use it. The second is working capital: SRS Acquiom's study of 1,250 private-target deals finds buyers' working-capital calculations accepted in roughly 7 of 10 cases, and contested ones typically resolve within two months through independent-accountant arbitration, costing $15,000 to $40,000 in fees split between the parties.

Financing type also moves the odds. On general lower-middle-market data, SBA-financed deals under $1M fail at rates up to about 40 percent, while cash-rich strategic buyers close at closer to an 85 percent rate. So who is buying you tells you a lot about your probability of getting to the finish line: a well-capitalised regional consolidator is a far surer close than a first-time searcher stitching together SBA debt. If the LOI-failure and diligence risk is the part that worries you, the exit-readiness work is where you de-risk it before signing, and the MSP exit lifecycle maps the full path from first conversation to close.

What this means if you are selling

Put the four pieces together and the median deal has a clear shape, and it is not the one the trade press sells you. You are most likely a bolt-on, not a platform. You will most likely price in the mid-single-digit EBITDA multiples, not the double digits. A real slice of your headline price will sit in a seller note and an earnout that, statistically, pays a fraction of its face value. And there is roughly a one-in-three chance the first LOI you sign never closes, most often because your earnings did not survive diligence the way you presented them.

None of that is a reason not to sell. It is a reason to sell prepared. The three levers that move a bolt-on outcome are all things you control before you go to market: clean, defensible earnings so the QoE recast does not blow a hole in your price; recurring-revenue quality and client retention so the earnout actually pays and the multiple sits at the top of your band rather than the bottom; and a buyer process with more than one credible bidder so you are not negotiating structure against your only option. Do those three and you can pull a sub-$10M deal from the low end of every table toward the high end, and cut the odds it dies on the table.

I run growth for MSPs and cyber firms, and I write these pieces to map the parts of this market that the headline coverage skips, because the median deal is where almost every owner actually lives. If you are thinking about a sale, the natural next reads are the valuation multiples that set your number, the buyer types that price you differently, and the first 100 days after a bolt-on closes, where a lot of these earnouts are won or lost. If you want to talk through where your business would land in these ranges, that is exactly the conversation I like having.

Frequently asked questions