Real Failure or Safe Failure? Reading the Interview Answer

Most failure stories told in job interviews are safe failures. Interviewers ask the question because they assume a candidate who admits to failing is at least being honest. But the failure was chosen before the candidate walked in. In six studies published in 2007, Julia Levashina and Michael Campion found that over 90% of undergraduate job candidates fake to some degree in interviews, and the failure question is the easiest one to prepare for. You pick a failure that costs you nothing.

What is a safe failure?

A safe failure is one the candidate can tell without losing points. It tends to come in a few shapes.

Some aren't really failures. Say the project shipped two weeks late, and it shipped. Or "I can be too detail-oriented."

Some belong to someone else. For instance, the market turned, headquarters cut the budget, the predecessor left a mess. The thing did fail, but in the story the candidate is a bystander.

And some are too old, like something from the first year of a career that has little to do with the job being filled today.

I wouldn't call any of this lying. Levashina and Campion's paper, in the Journal of Applied Psychology, built a scale of the ways candidates manage their image in interviews. One of its four categories is Image Protection, and its sub-behaviors include omitting, masking and distancing. Across the six studies, with 1,346 people in total, over 90% of undergraduate candidates did some faking. For the kinds of faking closer to outright lying, the share ranged from 28% to 75%.

The samples were students. The number may be different for someone with ten years behind them. The direction is easy to believe, though. An interview is an exam, and people choose the answer that helps them.

So when I hear a safe failure, I don't think "dishonest" belongs on the scorecard. What belongs there is that this box has no evidence in it yet.

The one question I always ask

There is one question I ask in every interview. What was the work experience you struggled with most, what did you learn from it, and if you did it over, what would be different?

Candidates usually come prepared for the first two parts. A struggle can be picked for safety. A lesson can be memorized. In the four layers I wrote that the way to see someone's patterns is to walk through one real failure, and that I don't put much weight on the lesson they drew afterward, because it has usually been rehearsed.

The third part works differently. To say what you would do over, you have to continue from the situation you just described. Where it got stuck, which decision you'd change, what would follow from the change. A memorized lesson doesn't connect to any of that.

Compare two answers. "I learned how important communication is" fits after any failure story. "If I did it again, I'd report the delay in week two and wouldn't wait for the end of the month" fits after exactly one. The first is a lesson. The second is a do-over.

Four places to listen

I sort what to listen for into four things: whether the situation is clear, where the candidate's own scope began and ended, what would be different a second time, and whether those pieces hold together. A safe failure usually gets through the first. It starts to leak at the second and third.

The failure that wasn't one tends to produce something like "honestly, I wouldn't change much." Then it wasn't the hardest experience. The hard one stayed home.

The failure that belonged to someone else has an empty scope. Ask about the do-over and everything that changes is somebody else: a different market, a bigger budget, another boss.

The old failure often has little detail left in it. Who was involved and what they had to work with stay vague.

A safe failure survives "what was the situation" and stops at "what would be different"

The four things an interviewer checks, in order, when hearing a failure story, and where each kind of safe failure leaks. This is the author's interview practice; no study stands behind the sequence

1. 1The situation thenConditions, timing, who was involved, told in order. A failure from too long ago runs out of detail here.
2. 2The candidate's own scopeWhich part was theirs, which decision they made. When the failure belonged to someone else, this is blank.
3. 3What would be differentWhich action changes, and when. A memorized lesson doesn't reach this far. A failure that wasn't one gives "nothing much."
4. 4Whether it holds togetherPut the changed action back into the situation. Does the outcome really change?

A lesson can be memorized in advance. A do-over has to continue from the situation the candidate just described.

The fourth check is the one that gets waved through. Take an example. The situation was that headquarters cut the budget. The do-over is "I'd communicate earlier." Would earlier communication have brought the budget back? If it wouldn't, the do-over and the failure aren't connected. My reading is that the candidate hasn't thought about the question before and is assembling the answer on the spot.

I wrote about taking apart success stories in "It was all me": finding what a candidate really did. Failure is the mirror image. Telling a success, people credit themselves with the team's share. Telling a failure, they hand their own share to the environment. Both questions are after the same thing, which is where this person's scope really was.

Junior and senior candidates are heard differently

With a junior candidate I only look at expression. Can they explain what happened? I listen mainly for the order of events and the part they were responsible for. I don't expect numbers.

With a senior candidate I look for a retrospective. After it was over, did they go back through it on their own?

A senior person without one answers while thinking. Some blame the environment, some blame other people, some blame themselves. What they can't produce is what would be different a second time, or they produce something and the logic doesn't hold. They have never asked themselves the question.

Blaming yourself is, I think, the version most likely to be scored as a plus. "It was entirely my fault" sounds like ownership. But self-blame and a retrospective are different things. Self-blame ends at "I was wrong." A retrospective gets to which step was wrong and what replaces it.

The best answer lays out the situation in order, and the logic has no gaps.

I'll put my judgment bluntly. A senior person with no retrospective isn't a senior who can carry responsibility and lead a team. The reasoning is my own inference. Someone who runs a team sees small failures from their people all the time, and walking through those is the job. A person who never went back over their own hardest experience is unlikely to do it for anyone else.

Does reviewing your own work actually improve results?

It does, by about a quarter on average. In a 2013 meta-analysis in Human Factors, Scott Tannenbaum and Christopher Cerasoli pooled 46 samples covering 2,136 people. Teams and individuals who held debriefs performed roughly 25% better than control groups that didn't. To count as a debrief, the review had to meet several conditions. Two of them were that it dealt with specific events and that the participants worked out the causes themselves. Passively receiving feedback didn't count.

That study measured what happens when people review their work after doing it. It didn't measure interview answers. The step from "people who debrief perform better" to "people who can answer the do-over question are people who debrief" is mine, and no research covers it.

On the scorecard, a safe failure leaves the box empty. It gets marked as unverified and the question comes back in the next round. As I wrote in After the resume screen: what I look for in an interview, a dimension with no event behind it doesn't get a score, and failure is no exception. The same question seen from the candidate's chair is in Telling failure as evidence.

What an empty box means in the final round depends, in my view, on the hire. For a junior role it matters little, because there isn't much history to draw on yet. For a senior hire who will run a team, the empty box is the answer.

---