Three-Quarters of Portfolio Leadership Is Bought, Not Built

In February, Altrata published the largest study of portfolio company leadership yet assembled — more than 11,500 US and UK portfolio companies, 60,000 executives, roughly 1,000 US private equity firms and more than 8,000 general partners. One finding in it should have moved more diligence budgets than it did. Seventy-four percent of US portfolio company leadership roles are filled from outside the company. Public companies, running the same seats, mostly promote.

That is not a small stylistic difference between two ownership models. It is a structural statement about where portfolio companies get their leaders, and it has a cost attached that almost never appears in an entry model.

Run the clock, not the org chart

The useful way to read this study is chronologically, against the hold period, because every figure in it lands at a different point on the same timeline.

At entry. Altrata's framing is that average holding periods are now "stretching to five to six years or more," and that with the 2021 peak well behind, sponsors are focused on "operational value creation, margin improvement and disciplined execution." Longer holds change the talent question from a swap to a sequence. A three-year hold needs one right leader. A six-year hold needs a leadership team that is still right when the company is twice the size it was at close.

Year one and two. This is where the external default shows up as an actual cost line. 68% of US and UK portfolio companies hire at least one leadership team member annually, and the CFO seat is the most frequently hired of all. Not the CEO. The CFO — the person who has to convert an operating thesis into reportable numbers, and who tends to be the first seat where the gap between "ran finance at a founder-led business" and "runs finance under a sponsor" becomes visible.

Across the hold. Here is the arithmetic nobody puts in the model. If 68% of companies make at least one leadership hire per year, and the hold runs five to six years, the directional expectation is somewhere around three and a half to four leadership hires per portfolio company over the life of the investment — and roughly three of those four coming from outside.

That figure is directional and should be treated as such. "At least one" is a floor, not an average, so the true number is likely higher; and the annual rate is a cross-sectional finding rather than a longitudinal one, so it assumes a steady state that no single company experiences. But even taken conservatively, it means the typical sponsor is underwriting a company that will run three or four executive searches before exit, and pricing exactly zero of them at entry.

Nobody models four leadership searches into an entry case. Every portfolio company runs them anyway.

At exit. And then the number that reframes everything above it: US portfolio company CEO tenure averages 5.8 years, which Altrata notes is "closely aligned with typical five- to six-year holding periods and significantly longer than other C-suite positions."

Read that against the churn data and a specific picture emerges. The CEO seat is not the volatile one. In aggregate, the chief executive arrives near the beginning and is still there at the end — the tenure figure and the hold period are effectively the same number. The instability is underneath the CEO, in the functional seats, and most of all in finance.

Why the bench is empty

The obvious question is why portfolio companies buy leadership when public companies grow it. The uncomfortable answer is that they are not choosing to.

A leader is developed over something like six to ten years of increasing scope. A hold period is five to six. The development cycle is longer than the ownership cycle, which means that under the current model a sponsor can rarely be the beneficiary of its own bench-building. The rational response to that constraint — buy the capability, do not grow it — is exactly what the 74% shows.

The problem is that the constraint compounds. Every portfolio company that buys rather than builds is drawing from a pool that other portfolio companies also failed to build. And you can see that closing loop in the data: three-quarters of today's US portfolio company CEOs and CFOs have previously held a C-suite or non-executive director position at a PE-owned business. Nearly two-thirds have prior buy-side or sell-side M&A experience.

That is a self-referential market. The same executives circulate between sponsors, which has two consequences that pull against each other. It raises the floor — these people have done a hold before, they know what a QofE is, they will not be surprised by the board cadence. And it raises the price while narrowing the field, because the qualifying criterion has quietly shifted from "can run this business" to "has run a business like this for someone like us."

Those are not the same test. The second one is a credential, and credentials are the part of a candidate that is easiest to verify and least predictive of whether the person fits this company — the same inversion we documented in The 11% Résumé, now operating at the portfolio level.

The sector finding nobody expected

Altrata's most instructive result is the one that refuses to generalize.

In US technology portfolio companies, 70% of externally recruited leadership team members came directly from prior tech roles. In US business services portfolio companies, fewer than one-third of external leadership hires came from the same sector.

Same ownership model. Same pressure. Opposite hiring behavior. And Altrata's read is that business services simply draws on "a broader cross-industry talent pool" — that in that sector, sector experience is not the binding constraint.

This is worth dwelling on because it destroys the single most common shortcut in sponsor-backed hiring: the assumption that the relevant question is "who has done this in our industry." Sometimes that is decisive — in technology, apparently, seven times out of ten. Sometimes it is close to irrelevant, and screening for it eliminates the better operator in favor of the more familiar résumé.

There is no way to know which case you are in from the outside. It depends on whether the value creation plan is a transformation or a multiplication — whether the company needs someone to change what it does, or to do far more of what already works. Those two jobs call for different people, and the industry-experience filter answers neither question. It is a proxy that gets applied because it is cheap, in a decision where being wrong costs a year.

What this argues for

Three things, in ascending order of difficulty.

Underwrite leadership churn at entry. If the base rate says three to four leadership hires across the hold, put them in the model — fees, vacancy cost, ramp time, and the productivity trough of a functional seat changing hands. An empty leadership seat has a daily cost that most sponsors have never calculated, which we broke down in The Empty Seat. Four of them across a hold is not a contingency. It is a line item.

Use the bridge deliberately rather than reactively. The reason interim and fractional leadership keeps appearing in portfolios is not that sponsors love the model. It is that the external default takes months to execute and the value creation plan does not pause. Deployed on purpose, a bridge buys the time to run the search properly instead of running it under duress — the case we made in The Fractional Bridge for PE-Backed Companies.

Build the bench anyway, even though you will not own the payoff. This is the hard one, and the honest counter-argument is real: if the development cycle exceeds the hold, bench-building is a gift to the next owner. Except that it is not, entirely. The layer directly beneath the C-suite is where the 68% annual hire either becomes unnecessary or becomes a scramble, and the same structural gap shows up outside private equity too — we wrote about the missing middle rung and what removing it costs later. A company with a credible number two in finance does not need to run a CFO search at the worst possible moment in year four. It also exits better, because a buyer is purchasing a team, not a person.

The limitation of this study should be stated plainly: it is a snapshot of who currently sits in these seats and how they got there. It does not tell us how those hires performed, and average tenure "so far" for sitting CEOs is a censored measure — it counts people still in the job, not the ones who did not last. Nothing here proves that external hiring produces worse outcomes than internal promotion in a portfolio setting.

What it does establish is the shape of the market. Leadership hiring in private equity is continuous, external by default, concentrated in finance, drawn from a pool that mostly recycles between sponsors, and governed by a sector-fit rule that reverses completely depending on the industry. That is a hard environment in which to hire by pattern-match, and pattern-matching is what most of the search industry sells.

The alternative is not more candidates. It is knowing, before anyone is sourced, which of the two jobs this actually is — transformation or multiplication — who the hire reports into and how that person behaves when a quarter goes sideways, and what the seat has to look like eighteen to twenty-four months out, when the company is materially different from the one the job description describes. Get that wrong at the top of a five-year hold and you find out in year two, which is the most expensive pattern in private equity. Get it right and even the first ninety days behave differently — though as we have written, those ninety days can still undo a good decision.

Four searches per hold, three of them external. That is the base rate. The only variable you control is how well each one is understood.