How to find an academic mentor when you don't know where to start

We hear a version of this story constantly.

A researcher gets lucky during their PhD: a supervisor who explains things patiently, never makes them feel stupid for asking questions, and leads by example rather than by lecturing. Check-ins happen without being asked for. Years later, that researcher still describes it as the foundation their entire scientific identity was built on.

Then the same researcher moves into a new position, often a postdoc, and the experience flips entirely. A supervisor who is warm and complimentary in the beginning, almost overwhelmingly so, then becomes inconsistent, hard to read, and ultimately damaging once the researcher is already committed and has no easy way out.

That gap, between how a mentor looks in week one and how they actually behave over the following two or three years, is the entire problem. Nobody teaches researchers how to evaluate a mentor before committing years of their life to them. They are expected to just know, get lucky, or learn the hard way. Stories like this are exactly why we built RateYourPI: because if there were a way to check before joining a lab, a lot of researchers would run the other direction.

Here's what we wish more people knew going in.

You can't tell a good mentor from a bad one in a single meeting

The single biggest mistake young researchers make is judging a potential mentor off the first impression. First impressions are exactly what a manipulative mentor is often best at producing. Excessive early warmth isn't always a red flag on its own, but it isn't evidence of anything either.

Here's the rule worth keeping in mind: enthusiasm in a first meeting tells you almost nothing about what two years of supervision will actually feel like. Good mentorship, like anything worth having, takes time to build and prove out. This doesn't mean every warm, enthusiastic PI is hiding something. It means warmth alone isn't evidence. Researchers need information that doesn't come from the person trying to recruit them.

Use data, not vibes: how RateYourPI actually works

This is exactly the gap RateYourPI was built to close. It's a platform for researchers to look up a potential PI and read honest reviews about them, before committing years of their life to find out it was a mistake.

Here's how to actually use it:

1. Go to RateYourPI.com and search the PI's name.

2. Read the reviews. These are written by people who've actually worked under that PI, not marketing copy from a lab website, not a polished bio.

3. Check the overall score, which is built from six specific criteria: Mentorship, Work-life balance, Publication ethics, Funding stability, International student support, and Overall experience.

Those criteria matter because a PI can be strong in one and weak in another in ways that aren't obvious from the outside. A PI might be a brilliant, generous mentor but chronically underfunded, which means a project stalls for reasons that have nothing to do with the researcher's ability. Another might publish constantly but cut corners on ethics in ways that could attach to a student's name later. International students in particular face risks, like visa stability, support during transitions, whether a PI will actually advocate for them, that domestic students may never think to ask about. Looking at all six together gives a fuller picture than any single conversation will.

The point isn't to find a perfect score. Few PIs have one. The point is to walk in with eyes open instead of relying on a 30-minute interview and a gut feeling.

Don't talk to current lab members. Talk to former ones.

This is the piece of advice we most wish more researchers heard early, and it cuts against what most people will tell you.

The standard advice is "talk to people currently in the lab." Don't, at least not only. Current students have every incentive to be careful with what they say. They may need a recommendation letter from this exact PI in a year. They may worry that something gets back to the PI and changes how they're treated. Even well-meaning current students will often soften the truth, consciously or not, because they're still inside the relationship and still need it to work.

Former lab members have nothing left to lose, which means they have nothing left to protect. They've already gotten their degree, their letter, their next position. They can say what actually happened.

How to find them: check the lab's alumni page, if it has one. Ask the department, most have records of who graduated from which lab and where they ended up. Look at the lab website itself, many list former members with where they went next, which gives a name and sometimes a current employer to search.

A short, honest message goes a long way: explain that you're considering joining the lab, that you'd value five minutes of their honest perspective, and that you understand if they'd rather not respond. Most people who had a good experience will gladly say so. Most people who had a bad one will gladly warn you. Either way, you learn something a current student usually can't safely tell you.

Putting it together

No single source, not RateYourPI, not a former student, not a great interview, is enough on its own. Real evaluation comes from combining all three.

The approach we'd actually recommend: search RateYourPI first, before ever getting on a call. Know the lab's track record on mentorship, funding, ethics, and support before investing any emotional energy. Find