How to find a mentor

1. Who to ask

The right mentor for a specific problem is rarely the most impressive person you can think of. It is the person who solved your exact problem recently enough to still remember the details.

• One to two levels ahead. Someone five levels up remembers the strategy. Someone one step ahead remembers the actual message they sent, the slide that got cut, and the number they asked for.
• Same function, adjacent company. A designer at a 200-person startup can tell another designer at a 200-person startup what will really happen in the next review cycle. Someone in a different function is guessing.
• Recently in the slump you are in. People who switched fields, got passed over, or came back from a layoff in the last two years give sharper advice than people who did it a decade ago.
• Actually reachable. A person with 400 followers who writes about your exact problem will reply. A person with 400,000 usually will not.

Write down three names before you write a single message. If you cannot name three, you do not have a mentor problem yet — you have a sourcing problem, and section 4 is where to start.

How to choose the right mentor (criteria checklist)

A good mentor match is about fit to one decision — not prestige. Check these before you send the ask.

• 1–2 levels ahead on the exact skill or decision — Good looks like: they recently made the same move and can name the tradeoffs you face, not only the outcome.
• Still practicing — recent operator, not only speaker — Good looks like: they still ship, review, or hire for the work. Talks and posts are signals; they are not a substitute for recent practice.
• Gives candor, not only cheerleading — Good looks like: they ask what you tried, push back on weak options, and leave you with one clearer next step.
• Comparable context (same function or adjacent company stage) — Good looks like: a designer advising a designer, or a PM at a similar-size company advising another PM. Same-context advice transfers; cross-function advice often only inspires.

If someone fails two or more of these, keep them as inspiration — not as your primary mentor for this problem.

Three questions before you ask

Answer these before you send a 20-minute ask — or in the first minutes of the call — to catch mismatch early.

1. What decision am I trying to make in the next 30 days? If you cannot name one, narrow first: portfolio narrative, level negotiation, switch path, interview loop, or manager feedback.
2. Why this person, not a peer or a search result? Point to one thing they did, shipped, or wrote that maps to your decision. "They are senior" is not enough.
3. What would a useful no look like? "I am not the right person — try X" or "no capacity this month" is useful. If you need a full career plan from a stranger, the ask is too big.

If you cannot answer all three in a short paragraph, rewrite the ask before you hit send.

2. The five-line outreach

Do not ask someone to be your mentor. It is an open-ended commitment to a stranger, and the safe answer is no. Ask for one conversation about one problem. Five lines is enough:

1. Who you are — one line, role and stage, no career history.
2. The specific problem — narrow enough that they can picture it.
3. Why this person — the thing they did or wrote that made you ask them and not someone else.
4. The ask — 20 minutes, with a time window.
5. The easy out — one line that makes declining cost them nothing.

Hi Priya — I am a product designer two years in, moving from agency work to in-house.

I am stuck on how to present agency projects in an in-house portfolio review, where nobody knows the client context.

You made the same move last year and wrote about the portfolio part of it.

Could I ask you for 20 minutes in the next couple of weeks to look at two slides?

If it is not a good time, no problem at all — I will keep reading either way.

Send it to all three names. Expect one reply. That is a normal hit rate, not a verdict on you.

3. The first 20 minutes

Bring one decision, not your life story. The most common way a first conversation is wasted is spending twelve of the twenty minutes explaining background the mentor does not need.

• Two minutes of context. Where you are, where you are trying to get to. Practise it out loud once.
• Fifteen minutes on one decision. "Do I take the lead role at the smaller company or stay?" beats "how do I grow my career?" Show the real artefact — the portfolio, the offer, the roadmap.
• Three minutes on next steps. Ask what you would do first if you were them. Write it down while they are still on the call.

Afterwards, send three lines: thank you, the one thing you are going to do, and when you will report back. That last line is what turns a single conversation into an ongoing one.

Red flags

Skip or pause if you see these. One flag can be a bad week; a pattern means find someone else.

• Always unavailable / cancels — Repeated cancels, ghosted confirmations, or no concrete window means the relationship will not hold. Ask someone who can keep a short appointment.
• Lecture-only — no questions about your context — If they never ask what you tried or what decision is on the table, you are getting a podcast, not mentorship. Useful mentors diagnose before they prescribe.
• Pushes dependence or unpaid labor — "Message me every day," "join my unpaid project first," or turning you into free labor without a clear learning trade is not mentorship.
• Outdated practice for your field — Prefer people who still do or hire for the work recently over people who left the craft years ago and only speak in generalities.
• Hidden agenda (recruiting, selling, status) — A soft product mention is fine. A pitch deck, course pressure, or using your session mainly to recruit is not. Thank them and leave.

Match by career stage

Keep the first ask to one row's worth of help. Expand later; you cannot recover from an oversized cold ask.

4. If you do not have a network yet

Most mentorship advice assumes you already know someone. If you do not, work channels in order — warm first, platforms last.

Warm paths first. Reply with substance to people already posting about your craft. Shared threads, meetup follow-ups, and "I tried what you suggested" notes beat a cold "will you mentor me."

Communities, alumni, and your company. Craft Slack/Discord groups and niche newsletters are where operators gather. Alumni directories are underused — shared background lowers the cost of a yes. One team over at your company is often the cheapest mentor to reach.

Free mentorship platforms (one option among peers). When warm paths are empty, free platforms exist for a specific question and no network: community office hours, open mentor marketplaces, and volunteer 1:1 platforms. ADPList is one of those — mentors volunteer video calls at no cost to mentees. It is not the only route. The five-line ask in section 2 still matters more than which channel you used.

If it helps to see who is out there by craft:

You can also narrow by company — Google, Meta, Amazon — or by where someone is based, such as the United States or the United Kingdom.

5. Turning one conversation into a relationship

1. Do the thing. The single strongest signal you can send is coming back having acted on the advice.
2. Report back within two weeks. Short note, what you changed, what happened. No ask attached.
3. Only then ask for the second conversation. With progress in hand, the second yes is far easier than the first.
4. Pass it on. Once you are the person one step ahead, someone behind you is looking for exactly you.

Related guides

• Best Mentorship Platforms in 2026: Free & Paid, Compared
• Free Design Mentorship — Find UX, Product & UI Design Mentors
• How to Switch Careers — A Practical Guide for 2026
• ADPList in numbers — verified mentor, session, and country figures