When to start preparing a portfolio company for exit, and what to do first

When should exit preparation start?

Begin structured preparation 24 to 30 months before the date you want the teaser in buyers' hands. That is our working rule, not a market standard, and it puts two items first: a metadata-only records review and, if the CEO is interested, the decision on a data license. A license needs a decision by the company, a rights review and a signed contract, and it is far easier to explain to buyers as a finished fact than as a live negotiation.

The reason is sequencing more than effort. A quality of earnings provider reads trailing twelve months, so a pricing fix or cost program finished six months before launch shows in only half of that window. Management depth, IP assignments and clean contracts take quarters to put right. Starting late does not just compress the work; some fixes never reach the numbers a buyer sees.

Reported hold periods differ by measure and year, so work back from your own fund's plan rather than a median. PitchBook put the median holding period of PE assets sold in the first half of 2024 at 5.8 years, down from a record of about seven years the year before, while Bain's Global Private Equity Report 2026 puts buyout holding periods at exit at around seven years. On either figure, a 24 to 30 month runway opens somewhere in years three to five of the hold, often before anyone has formally decided to sell.

What you need before you start

• An agreed launch window. The deal team and operating team should hold the same target quarter, even if it later moves.
• A view of the likely buyers. Strategic acquirers, sponsors and continuation buyers probe different things in diligence.
• A named owner at the company. Usually the CFO, with a budget for advisers and time protected from the monthly close.
• The current value creation plan and scorecard. Preparation should build on what has been delivered, not restart the story.
• A one-page systems list. Names only, for the records review in step 2.

A month-by-month exit preparation schedule

1. About 30 months out: set the exit thesis. Agree in one paragraph what the next owner is buying and which three or four proof points must exist by launch. Everything after this step serves those proof points.
2. About 27 months out: run the records review. The CFO and IT lead list every system holding operational records, how far back each one's history reaches, who is able to export it and whether the company clearly owns the content. Nothing is opened or shared. Run the result through the company fit checker, a preliminary and non-binding screen.
3. About 24 months out: decide on a data license. Explore, park or rule out. If the CEO wants to explore, introduce the company to SourceX now, so qualification, the data inventory, pricing and buyer review all happen while the sale is still a long way off.
4. From 24 to 18 months out: build numbers that survive diligence. Tighten the close, document every EBITDA adjustment, reconcile revenue by customer and keep any one-time proceeds separate from run-rate earnings.
5. From 18 to 12 months out: close legal and people gaps. Collect missing IP assignments, flag change-of-control clauses, name successors for key roles and agree retention arrangements.
6. From 12 to 6 months out: assemble the evidence pack. Match every claim in the future CIM to a document, rehearse the management team and commission vendor due diligence where it helps.
7. The final 6 months: select advisers and open the data room. By now the records inventory, any license agreement and the disclosure schedules should already be drafted.

The exit readiness checklist turns steps 4 to 7 into owner-by-owner tasks, and sponsors screening several companies at once can log every step 2 answer in the portfolio company screening workbook.

Why settle the license decision at month 24?

Settle it before the sale process because the pieces of a license sit on different clocks.

• Some stages are outside the company's control. Qualification, the data inventory and price and terms depend on the company. Buyer review and buyer selection do not, and no buyer is certain to select the data. Starting early leaves slack for the parts that cannot be hurried.
• A finished contract is easier to disclose than a live one. A license signed before launch goes into the data room as a completed agreement and a line in the disclosure schedules. One negotiated while buyers are in diligence invites new questions and puts deal counsel in the middle of the process.
• Exclusivity outlasts the hold. The licensee usually receives sole AI-training use for an agreed term, and the next owner inherits whatever remains of it.
• Management gets evidence. When buyers ask what data the company owns, a completed inventory or a signed license is a better answer than an ambition. The guide on the exit story and the value creation plan shows how to place it in the narrative.

If a buyer does select the data, payment is a single all-in amount, typically within about 60 days of invoicing, so finance can plan for it as a non-recurring item. An early introduction commits the company to nothing; only its own signature on agreed price and terms does that.

Common mistakes in exit preparation timing

Example: a records review at month 27 (Illustrative)

Illustrative and fictional. A sponsor owns a 240-person IT services company in year three of its hold and targets a launch about 27 months away. At step 2, the CFO lists 14 systems: a support desk with nine years of tickets, a CRM going back seven years, two retired professional services tools still archived, plus email, Teams, shared drives, finance and HR.

The rights check finds that client configuration files must stay out, because clients own them, but the company's internal tickets, runbooks and resolution notes are its own. The CEO agrees to explore a license, and the operating partner makes the introduction. Whatever the outcome, by launch the license is either signed and disclosed or ruled out, and the records inventory is already filed in the data room for the buyer's technology review.

If the timetable moves

Exits slip. If the target launch slips by twelve months or more, the schedule simply restarts from step 4, and the guide to value creation during longer hold periods covers levers that need no sale. For companies in a fund near the end of its life, see options for tail-end portfolio companies.

If a sale is already under way, do not start a license negotiation without deal counsel; record the records inventory for diligence and leave the decision to the next owner.

Next step

Put step 2 on the agenda of one portfolio company that is 24-30 months from its target launch. Strong answers justify an introduction: register as a partner, then either enter the company in the referral form or pass the CEO your link to the SourceX company application. The program overview for operating partners explains how referrals across a portfolio work.

Partners earn 25% of the eligible platform fees SourceX actually collects from the referred company's licensing deals, capped at $100,000 per referred company. Rewards become payable only after the buyer pays and SourceX receives its fee; an introduction, meeting or signed agreement alone does not trigger payment, and no reward is guaranteed. Operating partners should check fund documents and their firm's conflicts policy before accepting a reward linked to a portfolio company.