How to evaluate a CEO's leadership

To evaluate a CEO's leadership, look past results to six things: whether the strategy is legible to people who have to act on it, whether bad news travels up, whether the executive team has been shaped on purpose, whether capital goes where the strategy says, whether the CEO's behaviour is consistent across audiences, and what they're building that outlasts them. Results tell you what happened. These tell you what happens next.

I'm writing this for board members, investors and anyone trying to form a view on a leader from outside. It's also the question I work on from the inside.

Why results aren't the answer

A CEO can post three good years on a cycle they inherited, a market that lifted, or cost taken out of a business that needed investment. And a CEO can post two bad years while fixing something that was going to break anyway.

So results are the starting point and not the assessment. The useful question is whether the performance is repeatable, and that lives in how the place is being led.

1. Is the strategy legible three levels down?

Ask someone two or three levels below the executive team what the company is trying to do and what it has decided not to do. If you get a clear answer, and especially if you get the trade-offs, the message is working. If you get the slogan, it isn't.

This is the fastest diagnostic there is and almost nobody uses it. Ambiguity at the top doesn't stay ambiguous: it turns into politics, slow decisions and unclear ownership, and all three are visible from outside if you ask the right people.

2. Does bad news reach the CEO early?

Find out when the CEO learned about the last significant problem, and from whom. If the answer is late, or from outside the company, you've found something more important than any number in the pack.

The test applies to you too, if you're on the board. A CEO who is told things late by their own team is usually a CEO who punished someone for telling them early.

3. Has the executive team been chosen, or inherited?

Look at who sits around the CEO and ask whether that group was assembled for where the business is going. A CEO three years in with an unchanged inherited team has either made a deliberate judgement or avoided a decision. Both are worth knowing, and they look identical from the outside until you ask.

4. Does capital follow the strategy?

Compare what the CEO says matters with where the money and the senior people actually went over two years. Strategy is a budget, in the end. A mismatch here tells you the stated strategy is a narrative.

5. Is the CEO the same person in every room?

The board sees one version. The executive team sees another. The organisation sees a third. Where those accounts differ sharply, something is being performed, and performance under pressure is where leadership tends to fail.

The thing to look for is consistency of standard rather than consistency of style. Does the CEO hold the same line when it's costly as when it's cheap?

6. What's being built that outlasts them?

Ask what decisions this CEO has taken that pay off after their likely tenure. If every move lands inside the current cycle, you're looking at a manager of the present, which can be the right appointment and should at least be a conscious one.

Questions worth putting to a CEO directly

If you get an hour with them, these produce more than a strategy review.

  • What have you decided not to do, and who disagreed?
  • When did you last find out about a problem later than you should have, and what changed after?
  • Who on your team would you not hire again today?
  • What would your predecessor say you've got wrong?
  • What are you doing that only matters after you've gone?

The quality of the hesitation tells you as much as the answers.

What this doesn't measure

Likeability, which is uncorrelated. Confidence in public, which is a skill and often a mask. And charisma, which tends to buy a leader eighteen months of benefit before the questions above catch up with them.

If you're evaluating a specific CEO

The framework above works from public information plus three or four conversations. If you're a board doing this properly, the part people skip is the one in point two, because it requires talking to people who don't report to the CEO.

Book a call