What 63 exits taught us about holding periods

Every investment model we use assumes an exit in five to seven years. It is a sensible assumption and, for our best companies, it has been consistently wrong. This year we went back through all 63 realized equity exits since 2011 to see what the record actually shows.

The short answer: our median exit came at 7.4 years. Our top decile came at 9.8. The companies that returned the most capital were, almost without exception, the ones we were tempted to sell earlier.

1. What we measured

We looked at every realized position across venture, growth and real assets: entry date, exit date, route to exit and gross multiple. We excluded credit, where holding periods are set by loan terms, and partial secondary sales below 25% of a position.

The dataset is small by academic standards and specific to one firm. We share it because it changed how we think, not because it proves a rule.

2. The best outcomes came late

Exits held beyond eight years produced a median 4.1x gross multiple. Exits inside five years produced 1.6x. Much of that gap is selection: good companies get held, weaker ones get sold. But even within companies we rated highly at year five, those we kept outperformed those we sold.

3. The exits we nearly made

Four of our ten largest outcomes received credible acquisition offers between years four and six. In each case, the board's decision to continue was close. We now review every inbound offer against a written "hold case" prepared by a partner from a different strategy.

“The companies that returned the most capital were the ones we were tempted to sell earlier.”

4. What changes in our process

We are extending reserve periods in Venture Fund V, adding a formal hold review at year five for every growth position and reporting holding period data in our annual LP letter. None of this changes fund terms. It changes the questions we ask before we sell.

Conclusion

Patience is easy to claim and hard to practise when a fund is under pressure to return capital. The data gave us a reason to practise it more deliberately. We will report on whether it works.