Insights

What should a portfolio company do about AI in the first 100 days?

For pe operating partnersVerified July 2026

The first 100 days set the trajectory, and AI is now part of the value-creation plan whether or not it was in the deal thesis. The pressure is real — a board expecting a view, management with a hundred competing priorities, and a hold period measured in years, not decades. The temptation is either to do nothing yet or to launch a sweeping transformation. Both waste the window. The first 100 days are for a diagnosis and a few honest wins, not a moonshot.

This is how to use them.

Why the first 100 days

Attention and mandate are never higher than at the start of a hold. It is the moment when a new plan is expected, budget is being set, and the organisation is braced for change anyway. Spend that window on a genuine read of where AI can create value — and a couple of visible early wins — and you set an agenda the rest of the hold can build on. Let it drift, and AI becomes a standing "we should look at that" item that never gets a serious hearing again until it is too late to matter for the exit.

Don't start with a transformation

The instinct to announce an ambitious AI programme is the thing to resist. Big-bang transformations in the first quarter, before anyone understands the business's data, processes, or people, are how budget gets spent on science projects that never reach production. Start instead with a diagnosis: where does AI actually move a number here, what is the business ready for, and where are the risks. Pair that with one or two low-risk wins that prove the value is real. Credibility earned early is what funds the ambitious work later.

The 100-day arc

A workable shape trades the moonshot for momentum.

Phase Focus Output
Weeks 1–4 Get literate and honest Leadership up to speed on what AI can and can't do; a baseline of where AI is already used, and where value might sit
Weeks 5–8 Benchmark and govern A read of AI maturity against the peer group; the governance and EU AI Act gaps identified; a shortlist of high-value uses
Weeks 9–13 Prove and plan One or two quick wins delivered against a baseline; a costed, board-ready roadmap for what to fund next

The arc deliberately front-loads understanding and back-loads commitment. You reach the funding decision with evidence — a benchmark, a governance view, and a proven win — rather than with a slide deck and a hope.

What to put in front of the board

The output an investment committee can act on is not a list of tools; it is a roadmap that reads like any other value-creation lever: the value at stake, the cost to capture it, the risk (including governance and regulatory exposure), and a specific ask. Presented that way, AI competes for capital on the same terms as everything else in the plan, which is exactly where it should sit. A board funds a costed decision far more readily than an enthusiasm.

Portfolio-wide leverage

The quiet advantage for a PE firm is that the first-100-days playbook is repeatable. The diagnosis, the governance baseline, the benchmark, and the roadmap follow the same shape at each portfolio company, even as the answers differ. Run it once well and you have a template — and a growing view across the portfolio of where AI is creating value and where it is stalling — that turns a company-by-company scramble into an operating discipline. That is where the compounding return sits, not in any single tool.

Taking a portfolio company from AI uncertainty to a costed, board-ready roadmap in six weeks — literacy, benchmark, governance, and a proven win — is exactly what Firestarter's accelerator is built for, and it runs the same way across a portfolio.

Sources and verification. This guide describes a general approach to AI in the value-creation window of a PE-backed portfolio company as of July 2026. Every business and hold thesis differs; treat the arc as a starting template to adapt, not a fixed prescription, and validate any governance or regulatory steps against the company's specific obligations.

  • Risk & security lead

    How is shadow AI affecting my company?

    Shadow AI, the use of unapproved AI tools by employees, has become one of the fastest-growing data-governance risks for mid-market firms. A risk manager's guide to detecting it and protecting the business.

    Read the guide

  • Risk & security lead

    Does AI train on my business data?

    Whether an AI provider trains on your business data depends almost entirely on which plan you buy. A risk manager's guide to the safe tiers across Anthropic, OpenAI, Google, and Microsoft.

    Read the guide

  • Risk & security lead

    Which AI is right for my company?

    A cost and risk comparison of AI from Anthropic, OpenAI, Google, and Microsoft for business: API and per-seat pricing, plus which fits a manufacturing, financial services, or software company.

    Read the guide

  • Risk & compliance lead

    Does the EU AI Act apply to my company?

    The EU AI Act reaches companies well beyond the EU, and several of its obligations are already live. A risk lead's guide to whether you are in scope, which tier your AI use falls into, and the deadlines that matter.

    Read the guide

  • Risk & compliance lead

    What does an AI policy need to say?

    A good AI policy fits on a page and gets followed. A risk lead's guide to the essentials — approved tools, the data line that matters most, disclosure, and accountability — without the forty pages nobody reads.

    Read the guide

  • Chief financial officer

    Why do AI pilots fail to pay back?

    Most AI pilots never reach production, and fewer still return value the board can see. A CFO's guide to why AI pilots stall short of payback, and the four things the ones that pay back do differently.

    Read the guide

  • Chief executive

    How do I build a board-ready business case for AI?

    A board funds a decision, not a demo. A chief executive's guide to the four things an AI business case must contain — value, cost, risk, and the ask — and how to size it so the board can act.

    Read the guide

  • Change & people lead

    Why do employees resist AI?

    AI adoption stalls on people, not technology. A change lead's guide to the four real reasons employees resist AI, why training alone doesn't move them, and what actually changes behaviour.

    Read the guide

  • IT & data lead

    Is my data ready for AI?

    Whether your data is ready for AI depends entirely on what you ask AI to do with it. An IT lead's guide to the four things that actually matter, the permissions trap that catches everyone, and where to start.

    Read the guide

  • Operations lead

    What is an AI agent, and what should we automate first?

    An AI agent does more than answer — it takes actions across steps and systems. An operations lead's guide to what agents actually are, where they help and where they don't, and how to choose what to automate first.

    Read the guide