How to answer tell me about a time you made a difficult decision

Choose a decision where somebody specific carried a cost, and name them. Decisions that were merely complicated resolve once the analysis is done and demonstrate nothing beyond diligence. The version that scores identifies the trade-off you could not avoid and what you did about the person on the wrong side of it.

What if my difficult decision turned out to be correct

That is fine, and it is not the interesting part. Say what the decision cost and who absorbed it, then say what you would sequence differently. Interviewers are not scoring the outcome, which usually depended on things outside your control, they are scoring whether you understood the price at the time.

Should I admit I got part of the decision wrong

Yes, and be specific about which part. Getting the call right and the delivery wrong is extremely common and reads as self-awareness rather than weakness. An answer with no error in it anywhere invites the interviewer to go looking, and they will find one you have not prepared for.

Tell Me About a Difficult Decision You Made

Why interviewers ask this

When an interviewer says tell me about a difficult decision you made, most candidates reach for a decision that was complicated and present it as a decision that was hard. Those are different things and the difference is the entire question. A complicated decision has many variables and resolves once you have gathered enough of them, which makes it a research task with a deadline attached. A hard decision is one where you had enough information, both options were defensible, and choosing meant somebody specific carried a cost you could name in advance. Interviewers ask this because the second kind is the kind that actually happens above a certain level, and because candidates who have only made the first kind describe them identically.

An answer that sounds fine

We were deciding whether to rebuild our reporting pipeline or extend the existing one for another year. It was a genuinely difficult call because the rebuild was six months of engineering time and the extension carried real technical debt. I put together a comparison of the two paths covering cost, risk, maintenance burden and time to value, and I sat down with each of the three teams who depended on those reports most heavily. In the end the numbers favoured extending for a year and revisiting, because two of those teams had roadmap changes coming that would have altered the requirements anyway. We went with the extension. It turned out to be the right call, and it bought us the clarity we needed on requirements before committing six months of engineering time to something that would have been built against the wrong spec.

Where it breaks

Read that again and look for the person who lost. There is not one, which is how you know it was a complicated decision rather than a hard one. The candidate assembled a comparison, consulted the affected teams, found that the analysis pointed one way, and followed it. That is competent work and it required no judgment beyond the willingness to do the work, because at no point did he have to choose between two things he valued and live with what he gave up. It is the shape almost everybody brings when asked to describe a situation where you had to make a difficult decision. It also dissolves under a single question. Difficulty in these accounts sits in the analysis almost every time, and in the consequences almost never, which is the reverse of how difficulty works in an actual job.

The follow-up

What made it difficult, the options or the people?

The people, honestly, though that is not the part I usually tell. One of our engineers had spent about eight months building the original pipeline and the extension meant we were keeping something she already thought was the wrong design. I told her in the same all-hands where I told everyone else. She heard it with twenty other people in the room, which, thinking about it now, is the thing I would change.

There it is, and it took a direct question to surface it. The decision that was actually difficult was not extend versus rebuild, it was how to tell somebody that the thing she had built and already disowned was going to be kept alive for another year. He got the technical call right and handled the human half of it by putting it in an all-hands, which is what people do when they want the conversation to be over rather than had. A hiring manager does not need him to have been perfect here, and would be suspicious of a version where he was. What they need is for him to know which half of it was hard, because the person who thinks the spreadsheet was the difficult part will keep delivering the other half in all-hands meetings.

An answer that holds up

We chose to extend our reporting pipeline for a year rather than rebuild it. The analysis was not the difficult part, because the numbers favoured extending fairly clearly once two dependent teams told us their requirements were about to change. The difficult part was an engineer who had built the original pipeline, knew its design was wrong, and had been waiting a year for permission to replace it. I chose the extension knowing it meant asking her to maintain something she had already argued against. What I got wrong was telling her in an all-hands rather than beforehand, one to one, which turned a decision she could have disagreed with into an announcement she had to absorb in front of the team. She left the company eight months later. I do not think the decision caused that, and I think the way I delivered it was part of a pattern she was reading.

The interviewer asks what he got wrong in the decision.

Not the choice itself. The sequence. She found out in a room with twenty other people that she was keeping a design she had spent a year arguing against, and five minutes in private the day before would have cost me nothing.

Practice this question

Say your answer to this question out loud, then get asked whether the options or the people made it difficult. Answer it now, free, no account.