8 min. read

An AI-enabled roll-up buys existing services businesses and installs AI itself to run them at higher margins. Investors are putting serious capital behind the model, including $1.5 billion of General Catalyst’s latest fundraise as of September 2025. Read as a bet, that money says the companies already in those markets won’t build the capability themselves.

For a mid-market or PE-backed company, that’s a risk before it’s an opportunity: a competitor that may soon run at lower cost, in a market the company thought it understood. The hedge is to own AI deployment capability before a buyer or a rival does it for you.

Key facts

$1.5B Share of General Catalyst’s latest fundraise (as of Sept 2025) put toward its AI-enabled roll-up (“creation”) strategy, across seven industries with plans for up to 20. Source: TechCrunch, Sept 2025.

2× margin General Catalyst’s stated aim: at least double the EBITDA margin of the companies it acquires. Source: TechCrunch.

$6T+ Annual revenue of U.S. service industries, which General Catalyst calls largely untouched by meaningful AI integration. Source: General Catalyst.

$576.5B Gartner’s forecast for worldwide AI services spending in 2026. Source: Gartner, Sept 16, 2026.

What an AI-enabled roll-up is

General Catalyst, one of the model’s most visible backers, describes it as an approach that “pairs applied AI with strategic acquisitions of service businesses.” The firm backs founders who build AI for a specific services industry, then funds them to acquire and operate businesses in it: legal services, IT managed services, customer service and property management among them. Its argument is scale: U.S. service industries generate more than $6 trillion a year, more than the software market, and remain largely untouched by meaningful AI integration.

The economics are the point. TechCrunch reported in September 2025 that General Catalyst had put $1.5 billion of its latest fundraise toward the strategy and aims to at least double the EBITDA margin of each company it buys. Others, including investor Elad Gil, have backed similar plays, and the same report gives Gil’s case for the model: owning the asset lets you transform it faster than selling software to it as a vendor.

Why it’s a bet against incumbents

There are two ways to make money from putting AI to work in a services business. You can sell the capability to the company, or you can buy the company and install the capability yourself. The roll-up money chose the second. Roll-ups also earn returns the traditional ways, such as consolidating fragmented markets, but in our reading the AI thesis is strongest where the companies already in the market don’t close the gap themselves.

Your entry could not be saved. Please try again.

We sent links to your email! You should have received an email from us already. If you did not receive, make sure you check your spam folders and add masterofproject.com to safe senders list to receive our emails.

In our reading, most companies are still on the buying side of the first model. Gartner forecasts $576.5 billion of worldwide AI services spending in 2026. Buying help is often the right first move. The risk is buying help instead of building capability, until a competitor that owns the capability is the one setting prices.

The model isn’t guaranteed to work, and that’s the opening. TechCrunch’s report also notes early signs the transformation may be harder than investors expect, including the hidden cost of low-quality AI output that someone has to catch and fix. The advantage goes to whoever builds deployment capability that actually works. In our reading, incumbents start that race with the customers, the data and the domain knowledge.

Three risks for a mid-market or PE-backed company

1. A competitor that may run at lower cost

If an AI-enabled roll-up is buying in your sector and hits its margin targets, it can price below you and still earn more. That’s a valuation question for a PE-backed company, not just an operating one: the exit multiple depends on whether buyers see the company as the disrupter or the disrupted.

2. Renting alone doesn’t build muscle

Renting deployment talent gets the first system live. When the engagement ends, the learning leaves with them. The pattern isn’t new. In 2000, Toys R Us signed a ten-year agreement under which Amazon would oversee customer service and order fulfillment for the toy retailer’s online store. The partnership was meant to run through 2010; a dispute began in 2004, a 2006 court ruling found that Amazon had breached the deal and let the companies end it early, and in 2009 Amazon agreed to pay Toys R Us $51 million to settle (AP via ABC News).

In our reading, the lesson for an operator is about dependency, not theft: in the years when online retail was being learned, much of that learning was happening on the partner’s side of the contract.

3. The mid-market can’t easily rent in the first place

The price signals for deployment skill are set by large budgets. Job posts for mid-to-senior forward deployed engineers, the vendor-side deployment role, list $300,000–$550,000 in total compensation (Perspective AI), and Bloomberg reported in May 2026 that Wall Street firms pay AI trainers $25,000 a day. In our reading, a company too big to be a startup and too small to rent a deployment team is the one most exposed. Our guide to build, buy or rent prices each option.

The hedge: one role in each company

AI-native beats AI-renting, at any size. The hedge doesn’t require a lab or a large AI team. It requires one accountable person inside the company who owns each AI system after the vendor leaves and builds the capability that stays: an AI project manager. Our AI project manager job description template defines the role, and who owns your AI after go-live covers the handoff it owns.

For an operating partner, the five questions below make a quick exposure check across a portfolio.

Exposure question	Why it matters
1. Is an AI-enabled roll-up or AI-native service firm already buying in this company’s sector?	Those buyers aim to run the same services at higher margins. If one is active in the sector, a lower-cost competitor may arrive before the next hold-period review.
2. What share of this company’s AI deployment work is done by outside teams?	Renting gets a first rollout moving. When the engagement ends, the learning leaves with them unless someone inside was building it.
3. Who owns each AI system after the vendor leaves, by name?	A system nobody inside owns stalls or drifts. The owner is where deployment capability accumulates.
4. Could the company afford to rent deployment talent at market rates for as long as it needs it?	Job posts for vendor-side deployment engineers list $300,000–$550,000 in total compensation, and Wall Street firms pay AI trainers $25,000 a day. Prices set by large budgets are hard to sustain at mid-market scale.
5. Is there at least one AI-skilled role on the company’s own payroll, building capability that outlasts any single tool?	AI-native beats AI-renting, at any size. One internal role is the minimum that keeps the capability inside the company.

One diagnostic

Name the portfolio company most exposed to an AI-enabled competitor in its sector. Who inside it owns AI deployment today?

If the answer is a vendor, the company is renting the capability its competitor is buying.

Master of Project Academy sponsors the nation’s first registered apprenticeship dedicated to AI Project Management, registered by the Maryland Apprenticeship and Training Council (MATC #2575, September 8, 2026) under the registered occupation Information Technology Project Manager. The registration was announced via Business Wire.

Apprentices are full-time employees of the participating company from day one, so the capability builds inside it. The AI Project Management apprenticeship explains how the program works.

Frequently asked questions

What is an AI-enabled roll-up?

An investment strategy that buys existing services businesses and installs AI to run them at higher margins, rather than selling AI software to them. General Catalyst describes it as pairing applied AI with strategic acquisitions of service businesses, in sectors such as legal services, IT managed services and customer service.

How much money is going into AI roll-ups?

No complete public count exists. As of September 2025, General Catalyst had put $1.5 billion of its latest fundraise toward the strategy, according to TechCrunch, and it says it has committed billions of dollars to it. Other investors, including Elad Gil, have backed similar plays.

Should a mid-market company rent or build AI deployment capability?

Often both, in sequence. Renting gets a first system live; building keeps the capability when the engagement ends. Mid-market companies face rental prices set for large enterprises, so one internal role that owns AI systems after the vendor leaves is usually the minimum.

What should PE operating partners ask portfolio companies about AI?

Whether an AI-enabled roll-up is active in the sector; how much deployment work outside teams do; who owns each AI system after the vendor leaves; whether the company can afford to rent deployment talent at market rates; and whether at least one AI-skilled role sits on its own payroll.

Build the capability before a buyer does.

The AI Project Management apprenticeship, registered under the Information Technology Project Manager occupation (MATC #2575), develops one of your own employees into the role, on your projects, as a full-time employee from day one.

See how employers join the apprenticeship →

About Master of Project Academy

Founded in 2012, Master of Project Academy has served 500,000+ enrollments across 180+ countries in project management training. It sponsors the AI Project Management apprenticeship, registered under the Information Technology Project Manager occupation (MATC #2575). Learn more at masterofproject.com/p/aipmapprenticeship.

Sources

Business Wire — Maryland Registers Nation’s First AI Project Management Apprenticeship, Sponsored by Master of Project Academy (Sept 16, 2026)

The Money Is Betting You Won’t: What AI Roll-Ups Mean for Mid-Market Companies