Interview questions to ask candidates

The problem in most hiring loops is not a shortage of questions. It is that four interviewers ask versions of the same three questions, each hears the same prepared answer, and then the debrief discovers that nobody tested the thing this role actually turns on. This page assigns the questions. Each stage gets the questions it is allowed to settle, and nothing is asked twice.

The questions, stage by stage

Four stages, each with one job. Read the note above each group before the questions, because the note is what stops the stage expanding to fill the hour it has been given.

Stage one: the screening call

Owned by whoever runs your screen. Its job is to settle facts and dealbreakers, not to assess depth, and it should end in a yes or a no the same day.

What are you looking for that you cannot get where you are?: What it tells you: The reason for moving, in their own words, before anybody has coached them on what your company wants to hear. Somebody being pushed rather than pulled will usually say so at this stage and never again.
Talk me through the last two moves on your CV and why each one happened.: What it tells you: Continuity. You are checking that it is one story. Dates that do not line up are worth clarifying here, cheaply, rather than at the offer stage when a records question turns into a confrontation.; Follow up with: Was there any overlap between those two?
What does the role you want next look like on a Tuesday?: What it tells you: Whether their idea of the job matches the job. Candidates describe titles when asked about ambition and describe work when asked about Tuesday.
What is your notice period, and what have you seen colleagues actually serve?: What it tells you: Real availability rather than contractual availability. The second half of the question gets a more honest answer than the first.

Stage two: the hiring manager conversation

Owned by the person the hire will report to. Its job is to settle whether this person can do the work at your scale, and it is the only stage allowed to say yes with reservations, provided the reservations are written down for the next stage to test.

Tell me about the work you are proudest of that nobody outside your team noticed.: What it tells you: It removes the incentive to pick the story with the biggest logo attached, which is where the actual craft usually is.
What did you own end to end, and where did somebody else pick it up?: What it tells you: The seam between their work and the team’s. People from large, well-staffed functions often have a much narrower seam than the title suggests. That is not a fault, it is a fit question.; Follow up with: What would you have had to learn if that person had not been there?
What is the hardest part of this kind of work that people outside it never see?: What it tells you: Depth, quickly. It cannot be answered from a job description, and anybody who has done the work for two years has an immediate and slightly weary answer.
What is the smallest team and the largest team you have done this in?: What it tells you: Scale transfers more easily in one direction than the other. Somebody who has only done this with a support function around them will feel its absence here, and you would rather know now.

Stage three: the panel or the working session

Owned by the peers who will depend on the hire. Its job is to settle the work itself and the way they handle disagreement.

Here is something we are deciding this quarter. What would you want to know before you had a view?: What it tells you: You are scoring the questions, not the answer. The order in which somebody reduces uncertainty is the closest you will get to watching them work.
Where do you think we are getting this wrong?: What it tells you: Whether they will say a difficult thing to a room they want to join. Deference here becomes silence in month six, when it is expensive.
Show me how you would do the first version of this.: What it tells you: A short working session beats a long take-home. Watch what they clarify, what they assume, and whether they say plainly when something sits outside what they know.

Stage four: the closing conversation

Owned by the decision maker. The assessment is finished by now. This stage exists to settle whether the offer will hold.

What would make you say no to us?: What it tells you: The objection while you can still act on it. Ask it again even if the screen asked it, because the answer changes once somebody has met the team.
What else is live, and when do you expect those to close?: What it tells you: Your timeline, and whether the offer needs to move now or can wait for the second finalist to finish their loop.
Assume you join and it goes badly. What happened?: What it tells you: The risk they already suspect. It is the most useful sentence in the whole loop and it is almost never asked, because it feels like inviting doubt. It is inviting information.

The duplicate question audit

Put the loop on one page. Down the left, every question anybody plans to ask. Across the top, every interviewer. Almost every loop turns out to have three people testing communication and nobody testing the thing that failed on the last hire into this role.

Every question has exactly one owner. If two people want it, the more senior gives it up, because the candidate will tell a better version of the story to the second person who asks and you will record the practice as depth.
Every competency in the profile appears at least once. A competency with no question is being decided on impression.
Anything a screen can settle is settled at screen: notice period, band, location, work authorisation, and any hard requirement of the role.
Anything that comes back unanswered goes to the next interviewer in writing before their meeting, not into the debrief afterwards.
Anything nobody can attach to a competency gets cut. That is usually where the fifteen-question loop came from.

What each stage is allowed to decide

The screen decides facts

Dealbreakers, availability, band, and whether the story is coherent. It is allowed to reject. It is not allowed to pass somebody through on enthusiasm, and it is not allowed to assess depth, because the person running it usually cannot.

The manager decides capability

Whether this person can do the work at your scale, with your constraints, without the support structure their last employer had. This is the only stage that may say yes with reservations, and every reservation has to be written down for the panel to test.

The panel decides the work and the working

What they actually produce, and how they behave when somebody in the room disagrees. Peers read this better than managers do, because peers know what the work costs on a bad week.

The closer decides the offer

Whether to hire was settled upstream. This stage settles what the offer has to contain, who else has a say in it, and what would make it fall over between signature and joining.

Two candidates, two question sets

Somebody who has done this job before

Somebody stepping up into it

Ask what they would do differently the second time, and why they did not do it that way the first.

Ask for the closest thing they have done, then ask what will be different here. Accuracy about the difference is the evidence. Confidence is not.

Ask what they inherited and what they built, because a strong record at a strong company can be either.

Ask who currently does the parts of this job they have not done, and what they have watched that person get wrong.

Ask about the failure. Anybody with a decade in a discipline has one, and refusing to name it is more disqualifying than the failure was.

Ask about the last time they were out of their depth, and what they did in the first week of it.

Ask what they would need from you, and treat a long shopping list as data about their last environment rather than about them.

Ask what support they are assuming exists here, then say plainly whether it does. Half of first-time step-ups fail on an assumption nobody corrected.

Why loops grow a round

Nobody sets out to run six interviews. Rounds get added because a stage was allowed to reassess everything the previous stage already covered, so each one ends underinformed about the thing it was actually meant to settle, and the fix always looks like one more conversation. Give each stage one job and the loop shortens without anybody having to argue for cutting it.

A round added for a specific unanswered question is a good round. Write the question in the invitation.
A round added because the panel is uneasy and cannot say why is a symptom of a profile that was never agreed.
A round added so that a senior person can meet the candidate is a courtesy, and it should be scheduled as one rather than scored as an interview.
A round added after the offer has been discussed will be read by the candidate as doubt, and it is where good candidates take the other offer.

When the pile is the problem

None of this helps if the people reaching your calendar were selected by a keyword match. A well-designed loop applied to a badly selected shortlist produces careful, well-evidenced rejections, which is an expensive way of discovering that the screening was the problem all along.

What this looked like on a real role

18 months

running, and the funnel is still improving

On one long engagement the reading and scoring of inbound sat with the function while the client’s own TA team kept every decision. The share of the shortlist that team chose to interview rose across the year, because each reject and select sharpened what the bar actually meant.

Read the Appnomic engagement in full

Before you add a stage to the loop, price the extra fortnight it will cost you: work out the cost of the vacancy.

How Continuity1 runs this funnel

The screening above is the job. These are the numbers it produces when a function owns it end to end, set against the published benchmarks for the same market.

1 in 3: Shortlisted candidates you meet who become the hire; Aligned engagements run nearer 1 in 2, distant ones nearer 1 in 10. The market takes about 180 applicants to make one hire, and that sifting lands on your team rather than ours.; Continuity1 tracked engagements
~3: Interviews your team sits in, per hire; Ashby puts technical roles at 17.6 interviews per hire across the whole process, up 52% since 2021. The rest of that load sits with the function, not with you.; Ashby talent-trends report
95%: Offers that close inside your stated band; 20 of the last 21. A flat fee earns nothing from an inflated offer; a percentage of CTC earns more.; Continuity1 tracked engagements
1 in 9: Accepted offers that ghost before joining; Indian employers report nearly 4 in 10 offers dropped. We lose 1 in 9.; nasscom community

Every brief becomes a success profile before sourcing starts, calibrated with the people who will manage the role. That calibration is the step most hiring skips, and it is why a shortlist either matches the job or matches the job advert.

You review a scored shortlist and make the calls. The filtering never lands on your calendar.

Share a role See the workings and the citations

Questions teams ask

What interview questions should I ask every candidate?

The ones tied to a competency in the success profile, asked identically to everybody at that stage. Keep the primary question fixed so answers are comparable, and let the follow-ups vary, because their whole job is to chase what this particular person said.

How many interviews should a candidate go through?

Fewer than most loops run. Counts rise when each stage is allowed to reassess what the last one covered, so the loop grows a round every time somebody feels underinformed. Give each stage one job, write the unanswered question into the next invitation, and the loop shortens on its own.

Should I send candidates the questions in advance?

Send the topics, not the wording. Telling somebody you will spend twenty minutes on a project they led produces a better conversation and costs you nothing, because your evidence comes from the follow-ups. Sending the exact questions just moves the interview to the night before.

Who should ask about salary expectations?

Whoever runs the screen, on the first call, against a band you decided before posting. Discovering a mismatch after four interviews wastes the candidate’s time and your panel’s, and it is the most avoidable way for a loop to end in nothing.

What do we do when two interviewers disagree?

Get both scores in writing first, then look at what evidence each of them actually has. Disagreement usually turns out to be two people testing different things and calling it the same word, which is a design fault in the loop rather than a tie to be broken by seniority.