The single decision that shapes everything else — and how boards get it right.

Every board makes hundreds of decisions. Only one of them decides the outcome of all the others. Choosing a chief executive sets the direction of the company for the next five to ten years, and the cost of getting it wrong is rarely a single bad quarter. It is lost momentum, lost people, and a strategy that quietly stops moving. Yet in far too many companies, the CEO search begins the week the current one resigns — which is precisely the week a board has the least time to think clearly.

Hiring a CEO well is not a recruitment exercise. It is a governance exercise. The organisations that do it best treat it as an ongoing responsibility of the board, not a project that starts with a vacancy and ends with a signed contract.

Start long before the seat is empty

The best CEO appointments are made by boards that have been preparing for two or three years. That preparation is not a secret list in a drawer. It is a live, reviewed process: a shared view of what the company will need next, a small group of internal people being deliberately stretched and tested, and a working knowledge of who is available outside the company if the internal answer is not strong enough.

A useful discipline is to review succession at least once a year, with the same seriousness applied to the annual budget. Ask a simple question in every review: if the CEO left tomorrow, who takes over on Monday, who takes over in six months, and who could be ready in three years? If the board cannot answer all three, that is the work to be done — not later, now.

Emergency succession deserves its own plan. A named interim, a clear handover of authority, and an agreed communication approach for staff, customers, lenders and regulators can be written down in a single page. It costs almost nothing to prepare and it protects an enormous amount of value on the day it is needed.

Define the job before you define the person

The most common mistake in CEO hiring is starting with a list of qualities — visionary, decisive, commercial — that could describe almost anyone. These words feel meaningful and decide nothing.

Start instead with the strategy. What does the company actually need to achieve in the next five years, and what are the two or three hardest things standing in the way? A business that needs to rebuild its balance sheet needs a different leader from one that needs to enter four new markets. A family business preparing for outside investors needs someone who can bring structure without breaking the culture that made the company work. A technology-led company facing a shift in its industry needs someone who can make expensive bets and live with the discomfort.

Once the strategic problem is clear, the specification writes itself. Translate it into a short scorecard: three to five outcomes the new CEO must deliver, with a rough timeline attached to each. Then set out the capabilities genuinely required to deliver them. Everything else is preference, and preference should never quietly become a requirement.

This is also the moment to be honest about what the company is not looking for. A board that cannot say what it is willing to trade away will end up searching for a candidate who does not exist, and settling for whoever interviews best.

Look inside first — but look properly

Internal candidates carry real advantages. They know the business, the customers and the informal decision-making that no handover document ever captures. Transitions are usually faster and less disruptive, and the message it sends to the wider organisation is a powerful one: leadership here is earned.

The risk is familiarity. Boards see internal candidates through years of accumulated impressions, most of them formed in a role that is not the CEO role. Strong operators are not automatically strong chief executives; the job changes from running a function to setting direction, allocating capital and carrying public accountability.

The way to manage this is to assess internal candidates as rigorously as external ones — same scorecard, same interviews, same external assessment, same reference process. It feels awkward to put a colleague of twelve years through a formal evaluation. Do it anyway. It sharpens the board’s judgement, and it gives the successful candidate a genuine mandate rather than the appearance of an inside track.

External candidates bring fresh perspective, new networks and a willingness to challenge things nobody internally will question. They also carry higher risk: they take longer to become effective, they may misread the culture, and their track record was built in a different context. A strong record at a large, well-resourced company does not always transfer to a mid-sized business where the CEO has no layer to delegate to.

The right answer is to run both routes seriously and let the evidence decide. Boards that announce an internal succession and then quietly run an external search damage trust in both directions.

Assess judgement, not polish

Interviews reward confidence and articulacy. The CEO job rewards judgement under pressure. These are not the same thing, and a good process is designed to tell them apart.

Move away from questions about experience and towards questions about decisions. Ask candidates to walk through the hardest call they have made, what they knew at the time, what they got wrong, and what they would do differently. Ask them to review the company’s actual position — under a confidentiality agreement, with real numbers — and set out what they would do in the first year and what they would stop doing. The quality of the questions they ask in that session will tell you more than any answer they give.

Look hard at three things that rarely appear on a CV. The first is how they build and keep senior teams: strong CEOs attract strong people and are comfortable being surrounded by them. The second is how they handle being wrong, because every CEO will be, publicly, more than once. The third is character. Integrity, self-awareness and the willingness to hear uncomfortable news are not soft considerations; they are the difference between problems that surface early and problems that surface in a regulator’s letter.

References are where most boards do the least work and where the most is available. Go beyond the names offered. Speak to people who reported to the candidate, to peers who competed with them, and to those who worked with them during a difficult period. Ask specific, behavioural questions rather than general impressions. The pattern that emerges across ten honest conversations is usually a better predictor than any interview.

Govern the process, not just the decision

A CEO search needs clear ownership. In most cases that means a small nomination committee of independent directors, with defined authority, a published timeline and a chair who can hold the process together. Large committees do not improve judgement; they dilute accountability.

The sitting CEO should contribute information but should not control the outcome, particularly when internal candidates are involved. Founders and major shareholders need a defined role too — their influence is legitimate, but it should be exercised transparently rather than through side conversations that undermine the committee’s work.

Confidentiality matters more than boards expect. A leaked search unsettles employees, invites competitors to approach your best people, and pushes strong candidates away. Agree at the outset who knows what, and keep the circle small.

Finally, guard against the pull of the impressive candidate. Search processes generate momentum, and after months of work there is a strong instinct to conclude. If no candidate meets the scorecard, the honest answer is to extend the search or appoint a credible interim. An expensive delay is cheaper than a wrong appointment.

Get the terms right and the alignment will follow

Pay design is a signal, not just a cost. The structure of the package tells the incoming CEO what the board actually values, whatever the strategy document says. Weight the long-term incentives towards the outcomes on the scorecard, set performance measures the CEO can genuinely influence, and make sure the time horizons match the strategy rather than the reporting calendar.

Deal with the difficult clauses before the offer, not after. Notice periods, termination terms, restrictive covenants and the treatment of unvested awards are far easier to negotiate while goodwill is high. Boards should also be clear about the non-financial terms: decision rights, board reporting expectations, and where the CEO’s authority ends. Ambiguity here is the root of most early conflict between a new chief executive and a board.

The first year decides the appointment

The appointment is not complete when the contract is signed. Most CEO failures are visible within eighteen months, and many of them trace back to a weak start rather than a weak candidate.

Build a structured onboarding that goes beyond introductions: early access to the top three layers of management, direct exposure to major customers and lenders, and a deliberate handover from the outgoing CEO with a clear end date. Agree what the outgoing CEO’s role will be afterwards, and be cautious about keeping them on the board — it complicates authority at exactly the moment the new leader needs it.

Set expectations for the first hundred days that favour understanding over action, then review progress against the scorecard at six and twelve months. A designated director as first point of contact gives the new CEO somewhere to take questions that are not yet ready for a full board meeting.

The point of view

Boards that hire CEOs well share three habits. They treat succession as continuous rather than episodic. They define the job from the strategy rather than from a template. And they hold their nerve when a search does not produce the answer they hoped for.

None of this is complicated. It is simply harder than it looks, and it is almost impossible to do well under time pressure. The work to start today is not the search. It is the preparation that makes the search unnecessary to rush.

This article is intended as general guidance on board practice and does not constitute legal, tax or remuneration advice. Specific appointments should be structured with appropriate professional input.