Coaching, mentoring, or a better manager: which one do you actually need?

People arrive at “I think I need a coach” from a lot of different places, and a fair share of them need something else. The words get used interchangeably — coach, mentor, advisor, sounding board — and because the market is unregulated, anyone can call themselves any of them.

So here is the distinction as I’d draw it, and, more usefully, which problems each one actually fixes.

The actual difference

	Mentor	Coach	Manager	Training / course
Brings	Their path	A method and outside judgement	Context and authority	Content
Works on	Their experience, applied to you	Your live artefacts	Your performance in this org	A curriculum
Relationship	Informal, usually unpaid	Paid, structured, time-boxed	Structural, not optional	Transactional
Fails when	Their path doesn’t generalise	You want reassurance, not work	They’re also your judge	You needed practice, not information
Good for	“How did you get there?”	“Why does this keep happening to me?”	“What does good look like here?”	“I don’t know what RICE is”

The most important row is the fourth. A manager cannot fully be your coach, because they also decide your compensation and your promotion. There is a category of thing you will never say to someone who writes your review, and it’s usually the category that matters most — the doubt, the thing you got wrong, the job you’re thinking of leaving for. That structural limit is the single strongest argument for having someone outside the company.

Which one fits which stall

“I don’t know how people get to Head of Product.” A mentor. You want someone two or three steps ahead who’ll tell you what the path looked like from inside. Mentoring is cheap, informal, and undervalued — most people who’d be good mentors are simply never asked.

“I know what to do and it isn’t working.” Coaching. This is the signature case. The gap isn’t knowledge — you’ve read the books — it’s that something in how you operate keeps producing the same outcome, and you can’t see it from inside because you’re the constant across every instance.

“I don’t know what good looks like at this company.” Your manager, and if they can’t tell you, that’s the actual finding. This isn’t a coaching problem; it’s a missing feedback problem, and buying coaching to route around an absent manager is expensive and treats the symptom.

“My roadmap is a queue of whatever sales asked for.” None of the above. That’s a structural problem with how decisions get made, and no individual development fixes it. Either the org changes or you do — and I’d be doing you a disservice by selling coaching as the answer.

“I’m about to interview and I’m rusty.” Practice, not coaching. Work through the questions and what each one tests with anyone who’ll push back on your answers. A few hours of that beats a month’s engagement.

“I’ve just been made Head of Product and I don’t know where to start.” Genuinely a coaching case, and one of the highest-return ones — because the job changes shape at that step and almost nobody is prepared for it. It’s also the case where an advisor or interim alongside you can be a better fit than sessions; worth working out which before you commit.

What good coaching actually looks like

Four things that should be true, and are a fair test of anyone you’re considering:

It works on your material. Your roadmap, your last spec, the email you haven’t sent. If sessions are case studies and general principles, you’re paying for a course delivered slowly.

There’s homework, and it’s checked. Insight in a session that isn’t applied before the next one evaporates. The work happens between sessions; the sessions are for correction.

They tell you things you don’t want to hear. A coach who agrees with you is an expensive friend. The value is in the disagreement, delivered in a way you can act on.

It’s time-boxed. Coaching should end. A relationship that renews indefinitely has usually become a comfortable habit rather than a development programme.

The red flags

• A framework being sold rather than applied. If every client gets the same model regardless of their situation, you’re in a curriculum.
• No opinion about when you shouldn’t hire them. Anyone who thinks coaching is the answer to every stall is selling, not diagnosing.
• Sessions with no artefact. If nothing is written down, nothing is being worked on.
• Vagueness about price. You should get a number in the first conversation.

What it should cost you, and produce

Cost varies enormously and depends on cadence and commitment length, so I won’t pretend there’s a market rate. What’s more useful is the return test.

Before you start, write down what would have to be true in six months for this to have been worth it. Make it specific: a strategy doc that survives an exec review, the promotion, a decision process my team follows without me in the room. If you can’t write that sentence, you’re not ready to buy coaching yet — and any coach worth the money will tell you the same thing rather than take the engagement.

Then check it at three months, not at the end. The point of a short feedback loop is that it’s short.

When you need none of them

Three cases, said plainly because nobody selling coaching says them:

• You’re six months into your first PM role. You need reps, not optimisation. Almost everything that feels like a skill gap right now is actually a context gap and will close on its own.
• You’re burnt out. Coaching will make it worse, because it adds work to a system that’s already over capacity. Deal with the load first.
• The problem is the company, and you’ve already decided to leave. Coach yourself through the job search and start fresh; you’ll get more from coaching in month three of the new role than in month eleven of the old one.

---

If you’ve read this far and the second row of that first table is the one that described you — you know what to do and it keeps not working — that’s the problem coaching is built for. If it was one of the other rows, take the other route. It’ll be cheaper and it’ll work better.