The evidence favours internal promotion on cost and retention, and it is not close. External hires command roughly an 18 to 20 percent pay premium over internal people moved into the same jobs, receive worse performance evaluations for their first two years, and leave at higher rates. The case for going outside is real but narrow, and it rests on capability the bench genuinely does not have.
That gap between what the research says and what organizations actually do is the interesting part, because external hiring has been climbing while the evidence against it has stayed put.
What the research actually shows
Matthew Bidwell's work at Wharton is the study most often cited here, and it is worth being precise about what it found. Comparing external hires with internal movers in the same jobs, external hires were paid 18 to 20 percent more, scored lower on performance reviews for their first two years, and exited at higher rates. They also arrived with more education and more experience. They were, on paper, the stronger candidates, and they still underperformed the people already in the building.
At the top of the house the pattern repeats. Externally recruited CEOs are around 84 percent likelier to leave within their first three years than internally promoted ones, and external CEO appointments carry roughly a 15 percent pay premium. The full sourcing for each of these, with the underlying papers, is on our succession planning statistics page.
The explanation Bidwell offers is unglamorous and probably right: an internal candidate arrives knowing the organization. They know who actually decides things, which relationships matter, and where the work really happens. An external hire spends their first eighteen months acquiring that, and pays for it in output while being paid more for the privilege.
Why external hiring keeps rising anyway
Despite all of that, external hiring is trending up at the most visible level. Spencer Stuart recorded 147 CEO transitions across the S&P 1500 in 2024, with 44 percent of new appointments coming from outside, the highest external share since tracking began in 2000. The Conference Board found the S&P 500 succession rate rising to 13 percent in 2025, with external appointments nearly doubling from 18 to 33 percent.
Two readings of that are available, and they are not mutually exclusive. One is that boards increasingly want a change of direction, and an insider is a poor instrument for signalling change. The other is less flattering: boards go outside because the internal option was never built. Average CEO tenure is falling, which compresses the runway for developing a successor, and a bench that needed five years of deliberate development does not appear in the eighteen months between a board getting nervous and a CEO leaving.
The second reading matters more for most organizations, because it means a large share of external hires are not a choice between two good options. They are the only remaining option, priced accordingly.
When an external hire is the right call
The research is not an argument for promoting internally by default. There are cases where going outside is plainly correct:
- The capability genuinely does not exist inside. A company moving into a regulated market, a new technology, or a different operating model may need experience nobody internal has had the chance to acquire.
- The mandate is to break with the current approach. If the strategy being replaced is the one the internal candidates helped build, promoting one of them sends a message the board does not intend.
- The internal candidate is close but not ready, and the seat cannot wait. This is a real situation, and the honest response is to hire externally and keep developing the internal candidate rather than pretend the gap does not exist.
- The bench is a single name. One internal candidate is not a choice. It is a default with a story attached.
What the evidence does argue against is going outside because nobody can say with confidence whether the internal candidates are ready. That is not a decision. It is the absence of one.
The decision is made years before the vacancy
The internal-versus-external question feels like it gets answered when a seat opens. It does not. It gets answered by whether anyone spent the preceding three years building someone toward it.
This is why the comparison is slightly misleading as usually framed. By the time a role is vacant, the organizations with a credible internal candidate and the organizations without one are not making the same decision under different preferences. They are facing different option sets. The first can weigh a known quantity against an external search. The second can only run the search.
Keeping that option open is what a succession plan is for: scoring candidates against what the role actually needs, tracking readiness as it changes, and closing the specific gaps that stand between a promising internal candidate and a capable one. Bench strength is the name for having more than one such person. For senior roles specifically, executive succession planning is the version of this work that runs on the longest timeline and therefore has to start earliest.
How to make the choice deliberately
A few things separate a real comparison from a rationalized one:
- Define the role before you look at candidates. What the next holder of this seat has to be able to do, written down before anyone's name is attached to it. Otherwise the criteria drift toward whoever is in front of you.
- Score internal candidates against that definition, with evidence. "Not ready" is a useful answer when it comes with a specific gap and a date. "Not ready" as a general impression is how capable internal people get skipped.
- Price the external option honestly. The premium is not just salary. It is search costs, a slower first year, and a materially higher chance of a repeat search inside three years.
- Say which reason you are using. Going outside for capability the bench lacks is a decision. Going outside because nobody assessed the bench is an outcome. Both end with an external hire, and only one of them will look defensible in the postmortem.
The organizations that get this right are rarely the ones with a policy about internal promotion. They are the ones who can answer, on any given Tuesday, who could hold each critical seat and what still stands in the way. Building that is slower than a search and it is the only thing that makes the search optional.