In your first 90 days as CEO, listen before you decide, make your message unmistakable before you make your moves, and accept that the organisation is reading you from the first hour whether you've said anything or not. The sequence matters more than the content.
And the work starts before day one, which is where plenty of new CEOs lose ground they never get back.
The gap between appointment and first day is the only quiet time you'll get for a year. Use it.
Two things to settle. The first is what you're walking into: read the last eight board packs, the last two years of results, the leaver interviews if you can get them. Not to form a plan. To know which questions are already live.
The second is harder. Decide what you will not do, whatever the pressure. Here's the thing about the top job: it applies pressure to make you someone else, and it does it faster than you expect. The leaders who hold together are the ones who worked out their own line before anyone leaned on it.
So write it down. One or two sentences about what you refuse to betray. You'll want it by week six.
The temptation is to arrive with a plan. Resist it, even if the board hired you for one.
In the first month your job is to locate the business, which means finding out where it actually is rather than where the papers say it is. Those are different, always, and the distance between them is your first real piece of information.
What to do:
The two questions I'd use: what would you fix if it were yours, and what are we pretending is fine? The second one gets better answers than it has any right to.
One warning. How you respond to the first piece of bad news in this window sets the price of honesty for your whole tenure. People are watching that specific moment. If it costs someone something to tell you, it won't happen again, and you'll spend the next three years finding out late.
By now you know enough. This is where you make the message clear, and clear means the trade-offs are visible.
Ambiguity at the top doesn't stay ambiguous. It turns into politics, slow decisions and nobody owning anything, because when people can't tell what you've chosen they start guessing, and the guesses compete.
What your message has to contain:
Then say it at three lengths. Thirty seconds for a corridor, three minutes for a town hall, thirty minutes for your executive team. If it survives compression to thirty seconds, it's a message. If it only works at thirty minutes, it's a deck.
Actually, let me put that more usefully: test it by asking three people in different parts of the business to repeat it back. What comes back is your actual message, whatever you think you said.
Now you act, and the organisation learns what you meant.
Pick two or three moves, no more. Each one has to be visible, consistent with what you said in the second month, and costly enough that people can tell it was a choice. A move that costs you nothing teaches nobody anything.
The move people watch hardest is a personnel one. If you already know someone on your executive team isn't right for where this is going, the 90-day window is the cheapest moment you'll ever have to act, because nobody yet reads it as a verdict on your judgement. Wait a year and the same decision costs you three times as much, in credibility and in time.
That sounds brutal written down. In practice it's kinder than the alternative, which is a senior person spending eighteen months failing in public while everyone around them knows.
Go back to the sentence you wrote in Phase 0 and ask whether you're still that person. The honest answer at 90 days is usually no, not completely, and noticing it early is the difference between a correction and a crisis.
This sequence is chapter 11 of my book, Reinvention at the Top, in more detail and with examples.
Insider strategies in your inbox every Thursday