The lists that come up for this are long, and the length is the problem with them. A list you have not thought about is a list you pick from at random in the last few minutes, while you are tired and the interviewer has one eye on the clock. A smaller set works better, and only if you know what you are listening for when the answer comes back. The question is the cheap part. Reading the answer is the part that pays.
Two things are happening in those last minutes. You are still being assessed, so a question that shows you were paying attention helps you. And it is the only stretch of the hour where information moves toward you rather than away from you, so a question whose answer could not change a decision you make is a turn spent on nothing. The four below are picked to do both jobs at once.
"What would someone in this role need to have delivered a few months in for you to be glad you hired them?"
This is the job description rewritten by the person whose opinion will decide your first review. A posting is written once and then sits there. This answer is current.
A good answer is specific and slightly uncomfortable. It names a system, a backlog, a customer, something that is currently going badly. Someone who has thought hard about the gap will tell you what the gap is.
A weak answer reads the posting back to you, or lands on getting up to speed and building relationships. That is not automatically a warning, but it means the work has not been defined yet, and undefined work tends to get defined later by whoever is loudest.
"What do people find hardest about joining this team?"
Every team has a friction: a legacy system nobody wants to own, a stakeholder who changes their mind late, an on-call rota, a reorganisation that is halfway done. You are not really asking whether the friction exists. You are asking whether they will tell you about it.
A good answer names a real one and then says what they do about it. That is a strong signal in itself, because someone who can describe a genuine difficulty in front of a candidate works somewhere that allows saying so.
A weak answer is nothing really, everyone gets on well. Either they do not know, which is a problem if they are going to be your manager, or they do know and have decided you can find out once you have signed.
"How does the work get reviewed, and by whom?"
The reporting line on paper and the way decisions actually get made are frequently different, and the second one is the one you live in. This question surfaces the difference without asking anyone to admit to it.
A good answer names people and a rhythm: who looks at the work, and how often. A weak answer gives you names with no rhythm, or a rhythm with no names. Either can mean your work will be judged by someone you have not met, against a standard nobody has written down.
"Why is the role open?"
A new role and a backfill are different jobs. New means part of your first year goes on proving the role should exist. A backfill means there is an existing shape you are being fitted into, and it matters whether the last person was promoted, moved sideways, or left.
The content of the answer matters less than how easily it arrives. Most answers here are ordinary and unremarkable. Hesitation is the thing worth noticing, and it is worth one gentle follow-up rather than a raised eyebrow.
The process questions that are not optional
Ask what the next step is and when you should expect to hear. It is not a filler question. A specific answer is a small commitment, and it gives you a date after which following up is reasonable rather than pushy.
Ask about the pay range at the earliest point the process allows, which usually means asking a recruiter rather than the hiring manager. Ranges are easier to discuss before either side has spent several rounds on the outcome. Work out what a number means to you before you hear it: our salary calculator converts between hourly, weekly, monthly and annual figures, so you are not doing arithmetic in your head while somebody waits for you to react.
What not to ask
Anything already answered on the careers page. Asking it does not show interest, it shows you did not look.
Anything that only flatters the interviewer, of the what-do-you-love-about-working-here kind. You will get a pleasant answer that tells you nothing, and it costs you one of the few turns you have.
And do not stack them. Two questions asked well beat five asked quickly. If the interview is running short, pick the one whose answer would most change what you do next, then ask about next steps.
What to do with the answers
What you just heard is the most accurate description of this job you are going to get, better than the posting, because the person who owns the work said it out loud and unrehearsed. Write it down before you lose it.
It is also reusable. If the role turned out to lean on something the posting underplayed, that emphasis belongs in what you send for the next role like it: our resume keyword generator pulls the terms out of a posting, and tailoring a resume to a job description covers what to change once you have them. An interview that ends without an offer still leaves you better informed than you walked in, as long as you asked for something worth knowing.