The Delivery Intelligence Premium

Every MSP owner I've talked to thinks they know what their firm is worth. They've heard the ranges — 5x to 10x EBITDA, maybe higher if the revenue mix is right. What almost none of them have internalized is that the way buyers value services firms is in the middle of a structural break. Not a cycle. Not a correction. A permanent repricing of what counts as a valuable asset in enterprise services. AI is the catalyst, but the shift is deeper than AI — it's about whether your firm generates compounding, reusable intelligence or whether it sells hours.

The firms on the right side of that line are beginning to trade at software multiples. The firms on the wrong side won't understand why their multiple didn't move. This repricing has already swept through contact centers, legal services, and residential services. It hasn't hit the enterprise application managed services market yet — $60B today and projected past $250B by 2034 — where MSP's implement and manage mission critical systems like Workday, ServiceNow, Salesforce, and Oracle for the Fortune 2000. And it will.

To understand the economics of that gap, I analyzed over 50 disclosed and estimable M&A transactions across these ecosystems from 2016 to 2026, drawing on valuation frameworks from leading advisory firms, public company filings, PE operating partner commentary, and investment theses from the funds and founders who built and exited these businesses.

Here's what the data says- and the one thing the market hasn't learned to price yet.

The three levers the market already knows how to price

Revenue composition is the single most consequential valuation driver. ClearlyAcquired's analysis of per-dollar revenue value across IT services transactions: managed services revenue is valued at $1.33 per dollar. Project services at $0.65. Product resale at $0.12. A dollar of recurring revenue is worth roughly twice a dollar of project revenue in an acquirer's model. For a firm with $5M EBITDA, two additional turns from crossing the 60%+ recurring threshold translates to $10M in incremental enterprise value.

The thresholds are sharp. Below 50% recurring PE buyers won't engage. 60%+ is the tipping point where Breakwater M&A's 2026 analysis shows a consistent 1-2 EBITDA turn premium over project-based peers with identical profitability. 75-80%+ is best-in-class — Lincoln International's threshold for premium pricing.

In quality-of-earnings analysis, the disaggregation is ruthless. QoE analysts give full credit to contractual recurring revenue. Repeat project work (managed services contracts) with established clients gets moderate credit. Ad hoc Phase X work is heavily discounted: project-heavy businesses face a 15-30% valuation reduction. A QoE adjustment reclassifying just $2M from recurring to non-recurring at a 6x multiple swings the valuation by $12M.

Alight Solutions proved it in practice. They sold their Professional Services segment — Workday, SAP, and Oracle consulting — to H.I.G. Capital for up to $1.2B at roughly 10x EBITDA, then kept the BPaaS platform. Post-divestiture, they raised their EBITDA margin target by 600 basis points and pushed recurring revenue to 93.2%. The project business was the drag. The recurring platform was the asset. The entity left standing was worth more per dollar of EBITDA than the whole ever was.

The multiple ladder tells the rest of the story: 80% project-based clears 5-6x EBITDA. A 40/60 recurring/project blend reaches 7-8x. 60%+ recurring hits 9-12x. 90%+ recurring commands 15-19x. Same profitability. Completely different enterprise value.

Proprietary IP is the second lever. Chris Barbin — founder of Tercera, a growth equity firm dedicated exclusively to cloud services backed by Trilantic North America's $9.7B fund and the former CEO of Appirio, sold to Wipro for $500M — published an IP Maturity Pyramid defining four levels: sales assets, accelerators, standalone products, and productized solutions. This was three years ago, weeks before ChatGPT launched and the LLM wave made all of this obvious. The framework has only gotten sharper since. Customers want packaged expertise. Vendors want partners that extend their platform. Investors want margins that look like software. Barbin's warning: the market is littered with services firms claiming to be product companies and doing neither well.

Kainos (LSE: KNOS) is the controlled experiment. Three divisions, same company, same Workday ecosystem. Digital Services at 38.4% gross margin. Workday Services at 54.7%. Workday Products at 74-77% — growing 24% year-over-year while services revenue declines. That 35-40 point gap within a single firm is the clearest proof of the productization premium available in public markets. Every product in the portfolio — Smart Test, Smart Audit, Smart Shield, the acquired Genie testing product — emerged from consulting. Consultants observed repeatable problems across hundreds of deployments and codified the solutions into software now sold through Workday's own global sales teams.

The margin ladder is consistent across every source: custom consulting at 25-35%, managed services at 40-55%, productized services at 50-65%, pure SaaS at 70-80%+. Each step correlates with higher margins and a step-function increase in the multiple applied to those margins.

Size and scale is the third lever. An MSP valued under $5M trades at a median of 0.7x revenue and 5-6x EBITDA. Get to mid-market — $20-50M in enterprise value — and you're at 1.1-1.4x revenue and 8-9x EBITDA. Cross into platform territory above $100M and multiples expand to 1.6-2.1x revenue and 11-15x EBITDA. This is the engine behind every PE roll-up: buy small at 5-6x, consolidate, exit at 12-14x. Sunstone Partners merged three ServiceNow boutiques into Thirdera and exited to Cognizant at an estimated 4x return in three years. BV Investment Partners backed GlideFast Consulting and sold to ASGN at $350M — 3.7x revenue, roughly 25x EBITDA, still the high-water mark for a disclosed pure-play consulting deal. Trinity Hunt Partners's Argano has made 25 acquisitions since 2020. The size premium is real but it's the most understood lever. Everyone knows bigger is better.

All three — revenue composition, proprietary IP, and size — matter. But they're table stakes now. Every MSP has heard "get more recurring revenue" and "build IP." The edge isn't in knowing these levers exist.

The fourth lever — and the biggest opportunity in the MSP market right now

There's a fourth valuation lever emerging. It's the one with the most whitespace.

No MSP or AMS firm in Workday, ServiceNow, Salesforce, or Oracle has been acquired at a disclosed premium specifically for having a compounding delivery intelligence layer — purpose-built infrastructure that captures delivery patterns, configuration decisions, and engagement outcomes across every project, converting unstructured consulting experience into a reusable, compounding asset.

That's not a caveat. That's the opportunity. The first firms to build this won't be competing for a premium that already has a price. They'll be setting the price.

Distyl (software-enabled services firm) built exactly this and went from $200M to $1.8B in under a year — a 9x repricing. Its proprietary engine embeds into enterprise systems and creates automated AI Routines. Revenue grew 5x in 2024, tracking 8x in 2025, reaching 120M+ end users. Investors call them the "SEAL Team Six of AI engineering" — a consulting firm valued like a software company because its platform compounds delivery knowledge.

IBM acquired Hakkōda, an IBM Company in April 2025 and explicitly cited its "asset-centric delivery model" and "generative AI-powered assets" as the primary value drivers. Not headcount. Not certifications. The methodology and the AI assets built on top of it.

Titan raised $74M from General Catalyst to embed AI at the core of acquired MSP operations — deploying autonomous agents to handle routine tasks, compressing what used to take weeks into minutes. The thesis isn't to sell AI to clients. It's to rewire how the MSP itself operates.

The economic logic is sound: AI makes consulting knowledge codifiable — turning tacit expertise into reusable logic. Once digitized, it can be productized. When AI compresses a 10-week project to six, cost falls 30-40% while the client still pays for outcomes. But the structural impact goes beyond margins. It's valuation reclassification. You stop getting priced on EBITDA like a services firm. You start getting priced on revenue like a software company.

The emerging formula: base multiple (6-8x EBITDA) + recurring revenue premium (+2-4x) + IP premium (+1.5-3x) + delivery intelligence premium (+1-2.5x). That last delta is the one almost nobody is capturing — which means it's the one with the most room to run.

The ceiling is proven. EPAM Systems peaked at 7.7x revenue. Globant hit 10.1x. Hitachi acquired GlobalLogic at 8x — $9.6B. Most enterprise app MSPs trade at 1-2.5x today. That 3-6x gap is the largest valuation arbitrage in the MSP market. It belongs to whoever builds the delivery intelligence layer first.

The innovator's dilemma — and why most firms will miss this

This opportunity doesn't exist because nobody's noticed AI. Every MSP exec has heard the pitch. Every conference has an AI track. Every board deck has an AI slide.

It exists because of the innovator's dilemma.

The entire MSP operating model for the last twenty years has been built on two levers: charge your clients more or pay your people less. You either make the engagement look more complex to justify more billing, or you move delivery offshore to widen the margin. Every playbook is some variation of those two things.

So when AI shows up, most MSPs sell it as a service line. They bolt "AI consulting" onto the website and keep running the same operation underneath. They don't use it to change how their own firm operates, because that would mean questioning the model that got them here. The firms that got successful doing it the old way are the least likely to see that the old way is the ceiling, not the floor.

The ones that will command the highest exits are the ones willing to change their business model entirely. To operationalize their firm with technology in a way that genuinely creates value. Not as a buzzword. As the infrastructure that makes every engagement faster, cheaper, and more defensible than the last.

The gap between 1-2.5x and 5-8x isn't a timing window. It's a willingness window.

What this means for your firm

Here's what I keep coming back to. Most MSPs have IP. They just don't know it. They have years of implementation patterns, configuration decisions, migration playbooks, and institutional knowledge that took a decade to accumulate. But none of it is structured. It lives in the heads of three or four senior people. It walks out the door every time someone leaves. It gets rebuilt from scratch on every new engagement. And because it's never been codified, documented, or productized, it doesn't exist in the eyes of a buyer. It's not an asset. It's a liability disguised as expertise.

That's the first and most immediate unlock. Structure the IP you already have. Get it out of people's heads and into systems, frameworks, and repeatable tools. The research is unambiguous: that single move shifts a firm from body shop multiples to platform multiples. Ocean Tomo Investments confirms up to 20% acquisition premiums for firms with strong, documented IP. Kainos proved it in public markets, turning consulting observations into software products with a 35-40 point gross margin advantage over the services business sitting right next to it. This isn't theoretical. This is the most well-documented valuation lever in the entire M&A advisory literature, and most MSPs have never pulled it.

The second unlock is what separates the firms that will define the next era from the ones that got lucky in the last one. Build the delivery intelligence layer on top of that structured IP so it doesn't just sit in a repository. So it compounds. So every engagement makes the next one faster, cheaper, and more defensible. So a junior consultant on their second project can deliver at the level of a senior architect on their fiftieth, because the system carries the knowledge, not the person. That's what Distyl built to reach a $1.8B valuation. That's what IBM paid for when it acquired Hakkōda, an IBM Company. That's what Titan ($74M raised from General Catalyst) is embedding into every MSP it acquires. The research puts this at another 1 to 2.5x EBITDA turns on top of the IP premium. Most firms are capturing neither.

Together, structured IP and delivery intelligence represent 2.5-5.5X EBITDA turns of enterprise value that the majority of MSPs are leaving on the table. Not because the opportunity doesn't exist. Because they're still playing a game with two levers when the market has moved to four.

Revenue growth is the single most predictive variable in services firm valuation. Not scale. Growth. A firm growing at 30%+ with 70% recurring revenue, structured IP, and an AI-powered delivery layer isn't a consulting firm in the eyes of PE. It's an emerging software platform that happens to have a high-touch services wrapper. The gap between 1 to 2.5x revenue and 5 to 8x revenue isn't a timing window. It's a willingness window. The firms that wake up, the ones that stop optimizing the old model and start building the new one, those are the firms that will capture the repricing. Everyone else will keep running the same two levers and wonder why their number never moved.

The market hasn't learned to price delivery intelligence yet. That won't last. Someone will build the layer, prove the economics, and set the comp that reprices every firm after them. When that deal closes, every MSP owner will look at their own multiple and realize the game changed while they were still playing the old one.