# Good Coach, Feel-Good Coach

Published: 2026-07-13
Canonical: https://polyviam.com/writings/good-coach-feel-good-coach/
Excerpt: Good executive coaches do more than create space. They bring judgement, challenge, intervention and accountability to improve business outcomes.

About once a week, I speak to a CEO or founder who either has a coach or has worked with one in the past, but is now looking for another.

One of my first questions is usually: "What made you leave your last coach?"

The answer is nearly always some version of: "We got on really well, but I didn't feel like it was actually moving my business forward."

I've seen this before.

There is a sickness in the coaching world. The sickness is not that coaches "hold up a mirror", "create space", or help clients leave a session feeling energised. All of those things can be useful. The sickness is that they have _become_ the product, rather than tools _within_ the product.

The product should be improved executive effectiveness. Reflection is one way of achieving that. So are judgement, challenge, intervention and accountability. Somewhere along the way, much of the coaching industry forgot the latter four.

This can be actively damaging because it creates the sensation of progress. The client feels clearer. They have articulated the problem more elegantly. They leave with more energy than they arrived with. But unless that clarity and energy produce a better decision, changed behaviour or useful action, they are worth very little.

You see the same confusion in how the industry measures itself. Ask how executive coaching demonstrates its value and the evidence leans heavily on client satisfaction: whether people felt the sessions were valuable, whether they'd recommend their coach. That measures the feeling. It rarely measures the business. An industry that assesses itself on how clients feel about the work will, unsurprisingly, drift towards making them feel good about it. One test runs through everything that follows: a coach who is doing the work says things concrete enough to be proven wrong.

This is not unique to coaching. The wider leadership industry has the same tendency: attractive, virtuous-sounding ideas like "trust your people", "empower them", "hire great people and get out of their way", "lead with vulnerability". These aren't _incorrect_ per se. But - apart from empirical evidence pointing to the opposite of this for what creates effective leaders (note: this is about results, not morale) - they are also conveniently difficult to connect to the actual performance of a company. Traits in effective leaders are often less flattering: a high tolerance for conflict, political skill, hard choices about where resources go, and the willingness to intervene when something is going wrong rather than hoping the team resolves it themselves.

I once coached the founder of an online marketplace whose company was months away from running out of money. He was close to giving up and had suspicions that his leadership team was not capable of turning the business around.

After four sessions, and after observing the leadership team directly as part of a hybrid coaching and consulting engagement, I told him plainly that I thought he needed to fire them.

He did, the business survived, and is now performing well.

Had I merely asked, "What do you think?" or suggested that we construct a framework through which he could reach the answer himself, the conclusion might eventually have been the same. But surfacing the answer slowly was inferior to supplying judgement quickly because the stakes and time pressure mattered.

He was not paying me to create the perfect pedagogical experience. His company was about to fail.

Sometimes people already know the answer but lack confidence in it. Sometimes they are too close to the situation to trust their own judgement. Sometimes the coach has seen the pattern before and can shorten weeks or months of deliberation.

There is a serious argument on the other side, and it is not the one coaches usually make. The strongest case for staying non-directive is not that clients should discover every answer for themselves as a matter of pedagogy. It is an argument from asymmetry. The coach has a fraction of the context. The coach carries none of the downside. And the coach is sometimes simply wrong. Hand an outsider in that position the steering wheel and you have created a confident, unaccountable executive who pays no price for being wrong.

That argument is right about the risk and wrong about the remedy. The answer to a coach who might be wrong is not a coach who says nothing - it is a coach whose view is concrete enough to be tested, argued with, and discarded when it fails. The decision must remain the CEO's. They have the context, they own the company, they live with the consequences. But there is a large and useful space between making the decision and refusing to have an opinion about it.

A good executive coach should, in my view, be a quasi-consultant. They should understand enough about the company, the people, the economics and the operating context to form a serious view. They should get into the weeds, help untangle the problem and disagree openly about the available courses of action.

Otherwise, what are they?

A therapist? Possibly a mindset coach? A confidant? Each may have value. But none of them is necessarily improving your ability to run a business, and it is worth being honest about which one you are buying.

I play a lot of tennis and have had a lot of coaches. Tennis is a useful analogy because, however influential the coach may be, the player makes the decisions and owns the result.

But tennis coaches do not merely hold up mirrors. They diagnose technique. They devise tactics. They design practice. They identify when the player is avoiding a weakness. They say what they think should change, and the player sometimes tells them they are wrong.

Andre Agassi won six of his eight Grand Slam titles under Brad Gilbert, whom he later called the greatest coach of all time. It was not a relationship without argument. Gilbert's instinct as a tactician was to go after an opponent's weaknesses; Agassi wanted to attack their strengths, on the logic that breaking what an opponent relied on most applied the greatest pressure. Gilbert initially resisted the idea, then came to regard it as genius. That is what a working disagreement looks like: the coach held a concrete, contestable view, the player pushed back, and the better answer won. A coach with no view to defend could not have had that argument, and Agassi would have been worse off for it.

The same should be true in executive coaching. You should sometimes argue with your coach about how to solve a problem. Potentially vehemently. A coach who never disagrees either has no independent judgement, is withholding it, or has allowed the relationship to become organised around your comfort.

Disagreement does not mean the coach is right. A useful disagreement often ends with the coach changing their mind. But the coach has to be willing to say something concrete enough that it can be challenged and found wrong. "I think you are treating a strategy problem as a personnel problem" can be tested. "I am holding space for your journey" cannot. That is the line between a coach who is doing the work and one who has made themselves impossible to disagree with.

I also become emotionally invested in client outcomes. I get genuinely annoyed when a client repeatedly avoids a decision that we both know needs to be made, or fails to do something we agreed would materially move the business forward.

Some coaches would call that unprofessional. I think the opposite is more concerning. Emotional investment makes avoidance harder to ignore. A coach who can watch a client dodge the same decision week after week, keep charging for the sessions, and treat their own job as done once they've asked a thoughtful question, has incentives that are badly broken.

A CEO client and I once agreed that his number-one priority before our next session was a difficult conversation with his CTO. He wanted to move the roadmap back towards customer work, despite having promised the CTO a substantial period focused on technical debt.

At the next session, I asked how it had gone. It hadn't. "I was busy... there isn't much left on the technical-debt work anyway... I made a promise..." I let it slide. A week later, I heard the same excuses.

This time I was annoyed. I asked whether the CTO was free right then. He was. So I told the CEO we would reschedule our session and that he should go and have the conversation immediately. He did. It was resolved with minimal friction.

It honestly felt like telling a child to go and clean their room. But note what I was actually enforcing. I did not make the roadmap call for him, and I had no strong view on technical debt versus customer work. I held him to a decision _he_ had already made - that this conversation mattered and needed to happen. That distinction is the whole game. A coach who enforces their own preferences is the unelected executive I warned about earlier. A coach who holds you to your own stated commitments is doing exactly what you hired them for. The problem here was no longer a lack of insight. It was avoidance, and more reflection would only have bought him another week of it.

There is, of course, a balance between investment in the client's outcomes and professional neutrality. A coach who cares more about the outcome than the client does will eventually become controlling. But neutrality can also become indistinguishable from indifference; the same disengagement, just with better branding.

So it is not enough that your coach says they care. You should be able to see it in what they are willing to challenge, what they refuse to let slide, and whether they notice when you repeatedly fail to do the thing you said mattered most.

Leaving energised, getting on with your coach, having space to reflect - none of it is irrelevant. **It is simply not the product.**

A good coach may hold up a mirror. But when it matters, they also point at what is in it, tell you what they think it means, and argue with you about what to do next.