A Turnaround CTO for Portfolio Companies Is a Different Job

I get two kinds of calls about turnaround CTO work. The founder call starts with a story: what they built, where it hurts, what they're afraid of. The private equity call starts with two numbers. "Engineering is costing us $9 million a year and releases take a quarter. Fix both." Same title on the engagement letter. Very different job.

Most of what's written about interim technology leadership assumes the founder version. If you're an operating partner trying to source a turnaround CTO for a portfolio company, that writing will steer you wrong in a few specific ways. Here's what actually changes when the company answers to a fund.

The clock runs differently

A founder-led company lives on runway. A portfolio company lives on hold period, and the difference is bigger than it sounds. Runway is a survival constraint: don't die before the next milestone. A hold period is a return constraint: every quarter of bloated burn and slipped delivery comes straight out of the multiple at exit.

That math changes how a turnaround CTO should behave. In a founder company, three months of careful listening before making big moves can be wisdom. In a portfolio company with a five-year hold and eighteen months already gone, that same patience is a five-figure-a-day luxury nobody approved. The cost problem was visible in diligence. Every month it goes unattacked is EBITDA the fund already paid for and isn't getting. Sitting on it isn't caution. It's leakage.

So the tempo is different from day one. You still assess before you cut, but the assessment happens in weeks, against the model, with the deal team in the room. I've written about how I structure that period in my first 90 days inside a portfolio company.

The mandate is cost down, output up. Both.

Here's where the portfolio version separates hardest from the founder version. A founder usually wants growth protected at all costs. A fund wants the cost base attacked, and it wants delivery to improve while that happens, because the thesis priced in both. Most operators can do one or the other. Cutting is easy if you accept a slower company. Speeding up is easy if you accept a bigger bill. Doing both at once is the actual job.

The lever that makes both possible is AI, and I mean it concretely, not as a slide. When coding agents and copilots carry real work in the development loop, when AI handles the first pass of testing, code review, support and internal ops, a leaner team ships more than the bloated one did. That's what a top-tier AI-driven work environment is: the operating model rebuilt so AI is the default, with the delivery metrics to prove the machine got faster while it got cheaper. I've written about what that rebuild actually involves.

Sequencing is everything here. Cut before the AI leverage is in place and you've just made a slow company slower. Build the AI environment without touching cost and you've bought a science project. The two workstreams run together, each cut matched against the automation that absorbs the work, and the DORA trend watched weekly so nobody is guessing whether it's working.

The roadmap already exists. It's called the thesis.

In a founder company, part of the CTO's job is figuring out what to build. In a portfolio company, someone already decided, and they wrote it down, and a committee of people who do this for a living approved it. The investment memo is the roadmap. The margin assumptions in the model are the cost target. The technical work is in service of a document most engineers will never read.

This is why the first thing I ask a fund for isn't repo access. It's the memo. I've made this argument before in Read the Thesis Before the Code: architecture is only good relative to what the business is trying to become, and in a portfolio company that destination is unusually explicit. The same goes for cost. A cut that looks smart in isolation can be exactly wrong for the buy-and-build the fund actually underwrote. You can't tell from the invoices. You can tell in one read of the thesis.

The flip side matters too. Sometimes the thesis assumes savings the technology can't safely deliver on the timeline the model wants. A turnaround CTO who just salutes and swings the axe is as dangerous as one who protects every line item. The job includes walking into the operating partner's office and saying "this cut is real, this one needs the AI rebuild to land first, and this one was never true." Early, with evidence, while the plan can still absorb it.

The board speaks a different language

Founder boards tolerate a certain amount of technical narrative. Fund boards don't, and honestly, they shouldn't have to. When I report to a PE board I'm not describing engineering. I'm describing the deal: run-rate saved against the model's assumption, risks from diligence retired, and what the delivery metrics say about whether the leaner company can still ship what the thesis needs it to ship.

This is where measurement stops being optional. A founder might accept "the team is working hard." An investment committee wants a number that moved. It's why I lean so hard on DORA metrics in these engagements; they turn engineering health into a trend line a non-technical partner can read, and during a cost reset they're the tripwire that proves the cuts aren't breaking the machine. I've taken a portfolio company from low to elite DORA performance in ninety days, and the point was never the metrics themselves. It was that the board could see delivery improving while the bill shrank, instead of taking someone's word for both.

There are two clients, and one truth

The awkward part of turnaround work is that you serve two masters, and the cost mandate makes it sharper. The company needs a real executive: someone the engineers trust through a hard chapter, who isn't obviously the fund's axe. The fund needs honest reporting: what's actually true in there, not what the management team wishes were true.

The only way I know to hold both is to keep one version of the truth and say it in both rooms. The team hears the same assessment the board does, including about costs, translated but never softened into a different story. People can handle "we're going to run leaner and here's how the work still gets done." What they can't handle is discovering the real plan in a board leak. A turnaround CTO who maintains two narratives has become useless to both sides. Ask any candidate for this role how they handle it. If they haven't thought about it, they haven't done it.

What this means for sourcing

If you're evaluating candidates, the portfolio context sharpens the criteria. You want someone who has sat on the diligence side of the table and knows what a deal model looks like from the inside. You want an operator who has actually run an AI-driven organization, not one who read about them, because the rebuild is what makes the cuts survivable. And you want the handoff designed in from the start, because the turnaround CTO's last deliverable is a leaner, faster function the permanent hire can run without them.

I've written a separate piece on how to choose a turnaround CTO provider, including the firm-versus-individual question. The short version: you're hiring a person, whatever the packaging says.

If you have a portfolio company that's burning too much and shipping too little, that's exactly the work I do. Let's talk about the specific company →