Interview Questions to Ask Candidates (As an Employer)
The questions worth asking a candidate, and the part the lists leave out: what to do after they answer. Plus what you cannot legally ask, and how to interview for a job you can't do yourself.
·
There is no shortage of interview question lists. Search for one and you will find fifty, sixty, sometimes a hundred questions, sorted into neat categories. Most of them are fine questions. Almost none of them will tell you what you need to know.
The reason is simple. Every one of those lists is also read by the people you are interviewing. "Tell me about a time you handled conflict" has been answered in advance by every candidate who spent twenty minutes preparing. You are not testing whether they can do the job. You are testing whether they did their homework.
This guide gives you the questions worth asking. It also gives you the thing the lists leave out, which is what to do after they answer.
The only rule that matters: the follow-up is the test
A prepared question gets a prepared answer. A follow-up cannot be prepared, because it depends on what they just said.
Picture a normal exchange. You ask about a project they led. They tell you they rebuilt a broken process and the numbers improved. It is a good answer, delivered smoothly, and every question list would score it as a pass.
Now follow up. What was the number before? Where exactly were people getting stuck? What did you change first, and what happened? What did you try that did not work?
Someone who did the work answers all four without slowing down, because they lived it and the details are just sitting there. Someone who was adjacent to the work, on the team and in the meetings but not doing it, gets vague by question two and starts saying "we" instead of "I" by question three. The opening question produced a clean answer and no information. The follow-ups produced the information.
So treat the questions below as openings, not as a script. The list gets them talking. The follow-up is where you find out whether it is true.
Questions that test what is on their CV
These matter most, and most interviews skip them. A CV is a list of claims someone wrote about themselves. Treat each claim as something to check rather than something to accept, the same principle behind screening on evidence instead of keywords.
- Walk me through the project on your CV you are proudest of. What was your specific part of it?
- You listed [specific tool or skill]. Tell me about the last time you used it and what you were trying to do.
- Your CV says you [specific claim]. What did that look like day to day?
- What part of that result would not have happened without you?
- Who else worked on it, and what did they do?
- What went wrong in that project, and what did you do about it?
That last question earns its place. People rehearse their wins. Almost nobody rehearses their failures in detail, so the answer tends to come out unguarded. A candidate who cannot name a single thing that went wrong on a project they say they ran either was not running it, or is not being straight with you.
Watch the pronouns while they answer. "We shipped it" is a different claim from "I shipped it," and people slide between the two when the ownership is thinner than the CV suggests. On its own that proves nothing. Treat it as a prompt to ask one more question.
Questions that show you how they think
Your next problem will not be one they have already solved, so what you need to see is the judgment behind the finished work, not the finished work itself.
Ask them about a decision they made without enough information, and how they made it anyway. Ask what most people in their field get wrong. Ask about a time they changed their mind on something they had been confident about. Then put a real problem from your own business in front of them and ask how they would approach it in month one.
One more, worth asking every time: what would you need from me to do this job well? Strong candidates answer concretely. Access to a system, context on a decision, a tool bought, a meeting they should be in. Weak candidates say "just clear direction," which usually means they have not thought about it.
Questions about how they actually work
Your first few hires sit close to you, so how someone works matters nearly as much as what they can do.
- What does a good week look like for you?
- How do you prefer to get feedback?
- When you are stuck, how long do you sit with it before asking someone?
- Tell me about a manager you worked well with, and one you did not. What was the difference?
- What kind of work drains you?
Take these answers at face value. Someone who says they need structure has told you something useful about how to manage them, not something disqualifying. The mistake is treating this section as a personality test with correct answers.
Worth being honest about the limits here. Whether "culture fit" questions predict anything real is genuinely contested. The research is mixed, and the term gets used to justify hiring people who resemble the person doing the hiring. I ask these questions anyway, because working closely with someone for a year is a real consideration and pretending otherwise is not useful. But I would not let this section outweigh the evidence from the section above it, and anyone who tells you they can read culture fit reliably from a forty-minute conversation is overstating what they know.
When you are hiring for a job you cannot do yourself
This is the hardest version of the problem and a common one: hiring outside your own expertise. A non-technical founder hiring a developer. A technical founder hiring a marketer. You cannot judge the answer directly, because you do not know the subject.
You still have two tools.
The first is to make them explain it to you. Ask them to walk you through a technical decision in plain language, as if you did not work in their field. This works because clear explanation requires real understanding. Someone who knows the material can strip out the jargon. Someone repeating what they read cannot, because the jargon is all they have.
Three questions do most of the work here:
- Explain [something from their CV] to me like I do not work in your field.
- What is a common mistake people make in this area, and why does it happen?
- If I hired someone bad at this job, what would I not notice for six months?
That third question is the strongest one in this guide. It asks a candidate to describe what you cannot see, which only someone with real depth can do, and the answer hands you a checklist for evaluating everyone else you interview.
The second tool is a second opinion. One person who knows the field, sitting in on one interview, is worth more than any question list. If you do not have that person, the verification problem described further down is exactly the gap you are trying to cover.
Questions you must not ask
This part is not optional and it is where first-time hirers get caught. In most countries it is unlawful to make hiring decisions based on protected characteristics, and asking about them in an interview creates real legal risk even when you meant nothing by it. The specific categories and the exact rules vary by country, and sometimes by state or region, so confirm your local law. Nothing here is legal advice.
As a general rule, do not ask about age or date of birth, marital or relationship status, whether they have or plan to have children, childcare arrangements, religion, ethnicity or national origin, disability or health conditions, pregnancy, sexual orientation, or union membership.
Most of these get asked by accident, as small talk. "Do you have kids?" while waiting for a call to connect is a friendly question and a legal problem.
If your actual concern is whether they can do the job, ask about the job. Not "do you have young children" but "this role needs occasional evening cover, is that workable for you?" Not "do you have a disability" but "are you able to perform the core duties of this role, with reasonable adjustments if you need them?" Ask about the requirement, never about the person.
Salary history deserves a separate mention. It is now restricted or banned in a growing number of places, and even where it remains legal it is a weak thing to ask. What someone was paid before tells you about their last employer's budget, not about their value to you. Ask what they are looking for instead.
How to judge the answers
Decide what a good answer looks like before you start interviewing. Write down, for each of the three or four things this hire must do, what a strong answer would contain. Then ask every candidate the same core questions.
Asking the same core questions is what makes comparison possible. A different, freewheeling conversation with each person tells you who you enjoyed talking to. That is the bias behind most bad first hires. You drift toward the person who reminds you of yourself, then call it instinct afterwards.
Take notes during the interview rather than after. Memory reshapes itself to fit the impression you already formed.
The problem underneath all of this
Everything above assumes you already have a short list of people worth interviewing. Getting to that short list is the part nobody writes a question guide for.
You post a role and a hundred people apply. Maybe two hundred. Every one of those CVs is a set of claims, and the only way to know which claims are real is to do what this guide describes: ask, then follow up. For one person running a business, doing that across a hundred applicants is not possible. So most people skip it. They scan CVs for the right keywords, pick the five who wrote best, and start interviewing. The filtering already happened on writing quality, and no question list can undo that.
This is where BestHire fits. You post the role and get one link that collects every applicant. It sorts them on what they actually did rather than on keyword matches, so someone who did the right work in different words does not get dropped for using different vocabulary. Then it does the part you cannot do alone. It interviews every applicant who fits the role, not a handful you picked out first. Each interview is generated from that person's own CV and asks follow-up questions about the exact claims they wrote down, which is the same follow-up logic this guide is built on, run across the whole pool. It reads only what the person says, never their face, voice, or accent. Then it ranks them by whose claims held up and returns a short list with the evidence attached: the line from the CV, and the answer that backed it up or broke it. It rejects nobody. You make every call.
Which changes what your interview is for. The questions above stop being a search for whether someone can do the job, because you already have evidence either way. You use a few of them to confirm what the shortlist showed you, and spend the rest of the time on fit, which is the one thing no tool should decide for you.
The short version
- Canned questions get canned answers. The follow-up is the actual test.
- Start from their CV. Every claim on it is a question you have not asked yet.
- Ask what went wrong, not only what went well. Nobody rehearses their failures in detail.
- Hiring outside your expertise? Make them explain it in plain language, and ask what you would fail to notice for six months.
- Ask every candidate the same core questions, and decide what a good answer looks like before you start.
- Never ask about protected characteristics. Ask about the requirement, not the person.
This is the interview half of a bigger process. The full sequence, from posting the role to making the offer, is in How to Hire Your First Employee.