OfferDrill.

Interview questions

The hard ones

Tell me about a time you failed.

This page helps you pick a failure that's real enough to be believed but doesn't raise doubts about whether you can do the job. You'll get the structure, the traps, and three sample answers you can adapt to what actually happened to you.

Why they ask it

They're checking whether you can look at your own work honestly. Everyone has failed. The person who says they haven't is telling the interviewer they either can't see their own mistakes or won't admit them — and both of those are expensive to manage on a team.

They also want to know what you do after things go wrong. Do you spiral, blame, go quiet? Or do you tell someone early, fix what can be fixed, and change how you work? That second pattern is what they're listening for, and it's most of the score.

There's a third thing underneath. They want to see if you can talk about something uncomfortable without getting defensive or falling apart. How you handle this question is a small live sample of how you'd handle a hard conversation with a manager or a client.

What not to say

"I don't think I've really failed at anything."

It reads as either dishonest or oblivious. The interviewer stops listening for content and starts wondering what you're hiding. You've also burned the chance to show the one thing this question is built to reveal.

The fake failure: "I care too much," "I took on too much work because I wanted to help."

Interviewers hear this weekly. It's a dodge wearing a costume, and they know it. You get marked as someone who manages impressions instead of answering questions — which makes them doubt everything else you said.

A failure so small it had no consequence — a typo, a meeting you were late to.

It proves nothing. If the worst thing you can find is a typo, either you've never owned anything that mattered or you won't be honest about it. Neither helps you.

A failure that goes straight to the core skill of this job, with no recovery.

"I missed every deadline on that project" for a project management role, told without what changed afterward, is just a warning label. The failure itself isn't the problem — leaving it unresolved is.

Blaming: "It failed because my manager wouldn't give me resources."

Even if that's true — and sometimes it is — leading with it tells them you'll describe them the same way to your next interviewer. You can name a real constraint. You just can't make it the whole answer.

How to build a strong answer

  1. Pick a failure with real stakes and real distance.
    Two filters. First: did something actually go wrong that mattered to someone — money lost, a deadline blown, a client unhappy, a project killed? If nobody would have noticed, it's too small. Second: is it far enough from the center of this job that it doesn't become the interview? A missed launch date is fine for most roles. "I lost a patient's chart" is not fine for a hospital job. Pick something real, then check it against the job description. Most people have three or four candidates once they actually think. Older is fine — two to five years back is a good window, because you've had time to see what changed.
  2. Say what failed in one or two sentences, out loud, without cushioning.
    Name the thing plainly. "I led a system migration that ran three months late and we had to roll it back." No throat-clearing, no five minutes of setup about how complicated the situation was. The context comes next and it comes short. The flat, direct opening is what makes people believe you — and believing you is most of the battle.
  3. Own your specific part. Not everything, not nothing.
    Say what you did or didn't do. "I assumed the vendor's timeline was firm and never verified it." "I didn't escalate when I first saw the numbers slipping." If there were outside factors, you can mention one, briefly, as context — then come straight back to your piece. The formula: one sentence on the situation, then "what I got wrong was." That phrase does a lot of work.
  4. End on what changed, with proof.
    Not a lesson. A change. "I learned communication is important" is empty. "Now I get a written timeline from every vendor before I commit to a date, and I've done that on every project since" is a change. Best of all is evidence it stuck — a later project where the new habit caught something. Spend about a quarter of your answer here, then stop. Ninety seconds total, and the last thing out of your mouth should be a behavior, not an apology.

Examples to adapt

These are shapes, not scripts. Say them in your own words with your own facts — anything in brackets is a detail only you know, and it has to be the true one.

Mid-career, a project you led that went badly.

About [when], I led the rollout of [the system or project]. I committed to a launch date based on what the vendor told me, and I never pushed to verify it. Around week six it was obvious we were behind, and I kept thinking I could pull it back, so I didn't flag it to my director until we were already [the actual amount] past the date. We ended up [what actually happened — delayed, rolled back, over budget]. The timeline slip was partly outside my control. Not telling anyone for a month was entirely mine. What I do now is set a threshold up front — if we slip more than a week, it goes in the status report that Friday, whether I've solved it or not. On [the next project], that rule caught a supplier problem early enough that we moved the date once, in advance, instead of missing it.

The failure is real and had consequences. He owns the part that was his — the silence — while naming the outside factor once and moving past it. The fix is a specific rule, and there's evidence it held.

Early career, one internship or first job, not much scale yet.

During my internship at [employer], I was given [the task] and I was afraid of looking like I couldn't handle it, so I didn't ask questions. I worked on it for about [the actual time] and turned in something that was built on a wrong assumption about [whatever it was]. My supervisor had to redo most of it that week. She wasn't harsh about it, but I could tell it cost her time she didn't have. What I got wrong was treating questions as an admission I didn't belong. Now when I get something new, I write down what I think the goal is and confirm it in one message before I start. It takes five minutes and it's saved me from going the wrong direction twice since.

She doesn't pretend to have run a department. The stakes are small but real — someone else's time — and she doesn't inflate them. The fix is a concrete habit anyone can picture her doing.

A failure involving people, not a project.

I managed [the number] people at [employer] and one of them, [role], was struggling. I told myself I was being supportive by giving her room. What I was really doing was avoiding a hard conversation. Six months in, her performance hadn't moved, and by then the problem was big enough that [what happened — formal review, she left, the team picked up her work]. She told me later she hadn't known there was a problem at all, and she was right — I'd never said it clearly. That's on me. Since then, if something's off, I say it in the next one-on-one, in plain words, and I write down what we agreed. It's uncomfortable for ten minutes and it's saved two people on my team who might otherwise have been managed out.

A management failure is high-stakes and shows genuine self-examination. He names the thing he told himself at the time, which is the kind of detail people can't fake. And the change has a payoff for the team, not just for him.

What comes next

Answer this one well and the interviewer usually goes here:

  • What would you do differently if you faced that same situation tomorrow?
  • How did your manager react when they found out?
  • What did your team say about it?
  • Is that a pattern for you, or a one-time thing?
  • Tell me about a time you had to give someone else bad news.
  • What's something you're still working on?
  • When did you know it was going wrong, and what did you do in that moment?

Say it out loud before it counts.

Reading a good answer and giving one are different skills, and the gap shows up in the room. Offer asks you this question out loud, on a clock, and tells you what landed. The first one is free.

Practice this out loud

One honest failure, said plainly and followed by what you changed, will do more for you in that room than any polished non-answer — so pick yours tonight, say it out loud twice, and let it be enough.