Outgrace, home
LeadershipQ&A primer

Questions to ask before hiring an executive coach

The research on executive coaching is thin. Here is what to ask a provider, how to set goals you can score, and how to judge the result without a control group.

This primer is for founders, chief executives and board members who are about to pay for an executive coach and want to know what they are buying. The one thing to remember is this: independent evidence on executive coaching is almost absent from the sources we reviewed, so a buyer has to judge the process rather than trust the pitch. What follows are the questions that do that work, with the research that exists and the places where it stops. Where we give advice, it is our judgement and not a finding, and we say so.

What does the research say about executive coaching itself?

Very little that a buyer can rely on. The sources our researcher reviewed contain no study of executive coaching, no provider-vetting standard and no criteria for choosing a coach. The only academic work touching coaching is a Purdue University dissertation by Reed Jordan Engel, and it studies wellness coaches, not executives. Treat any claim of proven returns from a provider as a sales claim until they show you the study behind it, and check that the study covers executives.

Does wellness coaching research apply to leadership work?

Only loosely. The Purdue dissertation surveyed 673 wellness coaches and, in a separate online survey, 824 clients, and it was built to help clients keep health behaviour changes going. It describes wellness coaching as new in the early 2000s and its effectiveness as inconsistent across studies. That tells you the field is young and uneven, which is useful to know. It does not tell you what an executive coach will do for a chief financial officer who is struggling to delegate.

Do a coach's personality, background or experience predict results?

The evidence disagrees with itself. In the first two parts of the Purdue dissertation, coach and client traits suggested that outcomes were shaped by who the people were. In the third part, a matched-case analysis of 64 coaches and 413 clients, the study found no significant demographic or coach effect on outcome variables. Our advice is to stop choosing a coach on chemistry or on the strength of a CV alone, and to judge the process the coach runs.

How much of the change is down to the coach at all?

Less than most buyers assume, on the study's own measures. According to the Purdue dissertation, coach, client and demographic factors together explained 17.1% of the variance in clients' change in confidence, and 13.4% of the variance in the percentage of wellness goal achieved. The factors measured left most of the variation unexplained. In our judgement, the executive's situation, the sponsor and the work between sessions carry a large share of the result, so the engagement should be designed around them.

What should I ask about a coach's credentials?

Ask which accreditation or credential the coach holds, which body issued it, what training it required and whether you can confirm it directly with that body. Our sources include no accreditation standards, so we cannot tell you which credentials are rigorous. That judgement belongs with someone who has checked the bodies concerned, and it is worth asking your HR lead or an independent adviser to do that check. A coach who cannot name the issuing body in one sentence has already given you a useful answer.

How do I set goals that can be measured?

Write three to five outcomes the executive should be able to show within six months. Examples include a decision handled differently, a named behaviour in meetings, or a shift in the view of two or three stakeholders. Measure the baseline before the first session, not after it. Avoid goals such as 'become a better leader', which nobody can score. A useful goal is one that a colleague could confirm or refute without the coach in the room.

Who should agree the goals?

At least three people: the executive, the sponsor who will judge the result, usually a line manager or the board chair, and the coach. The Purdue dissertation examined client goals and barriers as one of its factors, but the summary we have does not report what it found about them. We would not rely on that gap to skip the step. If the coach writes the goals alone, the result will be measured against the coach's own view of success, and that is not the same thing as the company's.

How do I judge outcomes without a control group?

You cannot isolate the coach's effect from a single engagement, and you should not pretend that you can. Put the same short questions to the executive at the start and at the end, and to two or three colleagues, then compare the answers. Treat self-report with care, because the Purdue dissertation relied on online surveys and self-report is a weak basis for proof. A sensible test is whether a colleague can name a specific change in behaviour without being prompted.

What does founder training research tell me?

According to the National Bureau of Economic Research, Working Paper 34755 by Bailey, Fehder, Floyd, Hochberg and Lee reports a randomised, multi-year, multi-site field experiment with innovation-driven entrepreneurs. The paper raises a training effect hypothesis: training could shut some startups sooner while improving outcomes for others. The study concerns founder training, not executive coaching, so its relevance is indirect. Its value to you is the question it raises, which is whether any development programme helps people recognise when to stop, and that question is worth putting to a coach too.

When is coaching the wrong tool?

This is judgement rather than research. Coaching suits a defined individual behaviour, such as how an executive runs a difficult meeting, handles a new team or gives feedback. It is the wrong tool when the problem is structural, such as an unclear reporting line, a poor hire or a strategy the board has not yet settled. Before you buy, write the problem in one sentence and check whether it sits with the person or with the system around them.

What belongs in the contract?

Set out the scope, the number of sessions, the start and end dates, what the sponsor will receive and in what form, what stays confidential, and the terms for stopping early. If the contract touches data protection, confidentiality or employment terms, have a lawyer read it before you sign, because those points depend on your jurisdiction and are outside what we can advise on. A clear exit clause protects both sides if the goals stall.

How do I know when to stop?

Set a midpoint review against the baseline you wrote at the start. If the measures have not moved, the choice is between changing the coach, changing the goals, or ending the engagement. Stopping early is a legitimate result, and it is usually cheaper than paying for a process that is not working. Agree in advance who makes that call, so that the decision is not left to whoever feels least comfortable saying it.

The mistake behind most of these

The mistake behind most of these questions is treating a coach's credentials, warmth or reputation as evidence that the coaching will change the executive. Those things may matter to the working relationship, and a buyer is entitled to care about them. They are not results. In the Purdue dissertation, coach, client and demographic factors together explained 17.1% of the variance in clients' change in confidence, and the matched analysis found no significant demographic or coach effect on outcome variables. A good reference tells you the coach is safe to sit in a room with. It does not tell you the executive will change.

Brookings and the Pew Research Center, both of which our researcher checked, publish no coaching, executive development or accreditation findings in the sources reviewed. Brookings states a commitment to research quality and professional ethics, and that is a statement about the institution, not evidence about coaching. A provider who cites either name as support is offering general credibility. Ask which study the provider means, who published it, and whether it covers executives. If the answer is a logo or a general claim about research quality, it does not answer your question.

This week, do three things. Write down the one behaviour or decision the executive must change, and the three measures that would show the change. Send that list to the provider and ask them to amend it, not to replace it with their own. Ask for the credential, the issuing body and a way to confirm both before the first session. A provider who welcomes the test is giving you a fair basis for judging them later, and one who resists it is telling you something about the engagement.

Then write the midpoint review and the end date into the contract before anything starts. Send the confidentiality and data terms to your lawyer, and ask your HR lead or an independent adviser to check the accreditation claims with the bodies that issued them. If a provider answers these points in plain terms and accepts a measured baseline, you are buying a process you can judge. If they answer with a promise of transformation, you are buying a pitch, and no source we reviewed supports a promise of that kind.

Newsletter

The Briefing

One email a week: the leaders we profiled, the ideas worth stealing, and nothing else.

Weekly, free, one click to leave. Read our privacy policy.