SaaS Factory · 7 min read
How Private Equity legacy portfolio can become an AI era exit at a higher multiple
By Glyn Trott, — SaaS Factory CEOPrivate equity drove the last great enterprise technology wave, pushing cloud into portfolio companies over a decade ago, and the next wave is AI. So far, it isn’t paying.
AI spend is up across portfolios, but new net revenue is not. FTI Consulting’s 2026 Private Equity AI Radar, a survey of 200 fund and operating leaders, found that 95% of funds say their AI initiatives meet the business case, yet only 36% say portfolio companies use AI day to day, and just 7% describe AI as fully integrated. Plenty of promise, nothing landing in the P&L, and the spend is eating EBITDA while the team works out how to use it.
Meanwhile, the hold gets longer. Bain’s 2026 Midyear Private Equity Report puts average buyout holds at around 7 years, with roughly 33,000 unsold portfolio companies worth $3.8 trillion sitting in the backlog. McKinsey counts 16,000 companies held for more than 4 years, which is 52% of all buyout backed inventory, the highest on record.
The biggest funds are starting to get favoured treatment. Anthropic is reportedly in talks with Blackstone and others on a joint venture to embed AI across their portfolio companies, a Palantir style consulting model (CNBC, March 2026). The good news is you don’t have to wait for a joint venture with an AI lab, you can start working with SaaS Factory today.
SaaS Factory is built on one premise: speed to revenue, not just speed to code.
The old game was simple: buy at 8x, sell at 12x, and let the market do the work. With exits jammed, that game has stalled. A 2x return realised in 4 years compounds at roughly 19% a year, but the same 2x stretched over 8 years compounds at roughly 9%, and it’s simply taking years longer to exit (CapitalPad, 2026).
So value has to come from operations during the hold, and the delivery record is poor. Post merger research shows cost synergy capture runs at 70 to 85% of announced value, while revenue synergy capture is just 25 to 35%, and takes up to 3 years (Dealroom, 2026, citing BCG, McKinsey and Bain research).
And when the synergies stall, funds rework the model’s spreadsheet. The spreadsheet was never the problem, the software was. A software problem needs a software solution: products that work together across the portfolio, with overarching agents identifying the cross sell opportunity and then delivering it.
The instinct is to buy another tuck in and cross sell it. But a tuck in is a whole business, with its own ERP, its own contracts, its own people, landing on top of the integration pile you already have. The data is blunt on this: buy and build deals with 1 to 2 add ons earned 35.5% IRR (internal rate of return), while deals with more than 2 earned 19.9%, below standalone buyouts, and roughly 60% of roll ups miss their projected synergies within two years (CapitalPad, 2026, citing BCG/HHL research). More buying, worse returns.
The lower risk play is already sitting inside the portfolio. Your acquired customers pay other vendors for software and services every month. Go after that spend. Rip and duplicate the products they already use, but with fewer frictions, better integration with your suite, and AI on top. Their tech stack shrinks, their AI token costs drop, and your back to base net revenue climbs. It’s the share of wallet a tuck in would bring, without the deal risk, and without another business to merge.
Picture a platform with two tuck ins: 3 acquisitions, 3 core systems, nothing talks to anything. Replatform each one onto SaaS Factory and every product is built as part of a product suite, so product 3 lands already connected to the other 2, and everything you build after that connects too. The revenue synergy you underwrote stops waiting on an IT project. The products arrive pre connected, so cross sell starts on day one.
We proved it on ourselves. SaaS Factory core (the build platform), SaaS Factory Growth (strategy, channels and the built in CRM), SaaS Factory Revenue (billing, subscriptions and payments) and SaaS Factory Success (customer feedback into shipped software, the built in CSM) is not one monster app trying to be everything. They’re separate products, all connected, with cross product agents talking and working with each other.
And agentOS.com, the property CRM that started it all, shows the expansion pattern: blockmanOS as the Adjacent Market Product, Landlord Making Tax Digital as the Downstream Market Product, and Calmony Direct Debit collection as the ERP Market Product. One core product at the centre, and every product around it takes another slice of the customer’s spend.
Funds pay for what the pitch says, then spend year one discovering what the business actually is. The first quarter questions that matter most, real churn, real revenue per customer, real net retention, often can’t be answered because the acquired systems simply can’t produce them. And without those numbers, the year one pricing and upsell push is guesswork.
So the first 90 days are about visibility, and there are two truths to get. The commercial truth comes from the data: migrate it onto the platform and you can finally see real churn, revenue per customer and net retention, and price with confidence. The technical truth comes from the replatform: where the repo allows, SaaS Factory Replatform rebuilds fully specced, tested, quality checked and audited, with a SaaS Factory Quality Score, the state of the asset in one number.
Then the same data aims the build. This is the SaaS Factory Moat Score at work: Data Flywheel, Domain Depth and Workflow Lock-in. The signals in your customer data show you which Adjacent Market Product to build next, which Downstream Market Product your customers’ customers would pay for, and which connections would improve the ERP at the centre of it. By day 90 you’re not guessing what to build, you’re building what the data tells you to build.
Merge 3 companies and the product requirements are the same every time: billing and payments that work across entities, KYC and KYB, SSO and security, connections to the systems around it, customer feedback loops, growth channels being worked, and someone watching errors at 3am.
Historically that’s a team costing $80,000 to $120,000+ a month, per company. Every SaaS Factory product ships with all of it built in, AI agents, agentic flows and MCP from day one, with SaaS Factory Product Owner AI Agents running the SaaS business. The comparable running cost is $300 to $500 a month.
For a fund, the accounting matters as much as the number. Build costs are typically capitalisable (check with your advisers), so build sits on the balance sheet, and savings land in EBITDA, in the first quarter. The freed cash flow services the leverage, or pays back the funding faster.
Point the SaaS Factory Opportunity Engine at each portfolio company, or at the competitor your portfolio is looking to emulate or dominate. Within the hour you get a full report: what the company owns, its unfair advantages, and scored product ideas in two lanes, Adjacent Market Products for its existing customers and Downstream Market Products for its customers’ customers. [LINK: sample Opportunity Engine reports]
Then any one of the Adjacent or Downstream Market Product opportunities arrives already specced: instructions the SaaS Factory Product Owner Agent executes.
For fund CEOs, connect your AI to the SaaS Factory MCP and task your agents to risk assess and build out your quarter action plan.
Every buyer at exit now asks one question: could an AI native competitor rebuild this for 5% of the price? For a single legacy app, increasingly yes, and the decisions not to buy are growing.
The defence is the suite. Connected products, one MCP layer, agents running the workflows, built on the AI Era Defensible Moats: Domain Depth, Workflow Lock-in and the Data Flywheel. Then add Adjacent Market Products, Downstream Market Products that your customers distribute to their own customers, and ERP Market Products. Now your software doesn’t just sit inside the customer’s business, it runs through it and out the other side, and leaving would mean unpicking their own revenue. That is pleasantly complex.
Copying one app is cheap, copying the suite is not.
Buy legacy. Sell AI era.
Several Volaris team members told us private equity would want this, and they’d know. Volaris buys software companies from PE funds every year.
And being open, we have customers but have yet to work with a PE firm. If you like what we have written and would like to talk and potentially work with us, please reach out to glyn@saas-factory.ai
If your value creation plan is opex savings and a market recovery, we’re not for you. SaaS Factory is for private equity funds that intend to change what the portfolio can do and create speed to revenue.
Glyn Trott founded agentOS in 2004 and sold it to Volaris in 2024. He continues to be CEO of agentOS Proptech Group, part of the Omegro and Volaris group, and co-founder of SaaS Factory with Dylan Davies, CTO. Glyn lives in Wales and is big into CrossFit. Dylan lives in New Zealand and adores red wine. Glyn gets up at 5:30am to have his morning meeting with Dylan, Dylan gets up at 6:30am to have his morning meeting with Glyn. They have created a two person, 24 hour cycle company, supported by their product owner and business stack AI agent teams.
More from SaaS Factory
8 min read
Subscription, hybrid and outcome based pricing. AI is rewriting how SaaS companies charge, and staying just with subscriptions is not an option.
3 min read
Intelligence is no longer the bottleneck — every enterprise has capable agents. What has been missing is somewhere for that intelligence to safely build. SaaS Factory over MCP hands any agent the controls of an entire software factory: quality products, delivered and operated at a scale no team could match.
4 min read
We didn't copy what Lovable, Bolt and Cursor do brilliantly. We doubled down on what they all do badly. Here's what Reverse Benchmarking the AI build tools taught us, and why our only metric is the first pound in your bank account.
