‹ Field Notes
Customer focus

Underdog shipped a 32MB speech model inside a playable game. The question that tests whether "more" was ever the point.

Nobody orders a game with a model release. The expected deliverable is a weights file, a README and a benchmark table, and by that standard the room with the desk lamp is surplus. But look at what the surplus does. The claim a customer would be most sceptical of is the one in the company's own framing: that something this small really runs, offline, on hardware you already have. A benchmark asserts that. A typewriter that talks back in your browser demonstrates it, on your machine, before you have downloaded anything. Whatever Underdog's reasons, the demo reads as a particular kind of extra: not more model, but a sharper answer to the doubt the customer was going to raise anyway. That is the difference an interviewer is listening for when they ask you to describe a time you delivered over and above what a customer expected. The weak version of that story is about volume. The strong version is about aim.

The interview question
“Tell me about a time when you delivered over and above customer expectations.”

Why they ask it

The question sounds like an invitation to brag, which is why so many people answer it badly. What the interviewer is actually probing is whether you can locate a customer's real expectation precisely enough to exceed it on purpose. Anyone can add things. The skill is knowing which addition the customer will experience as "more than I hoped for" rather than "more than I asked for," and those are not the same feeling. The first builds trust; the second builds a backlog of features nobody uses.

They are also checking for judgement about cost. Exceeding expectations is never free. Someone spent time, scope or political capital, and a senior candidate is expected to know what they spent and to have decided it was worth it. A story in which the extra effort appears from nowhere, costs nothing and delights everyone is a story the interviewer will not believe.

The trap

The trap is the heroics answer. The candidate describes working through a weekend, shipping three features beyond the brief, or personally rewriting a component so it ran faster than anyone requested. The customer is pleased, the story ends, and the interviewer has learned that you confuse effort with value. Worse, the heroics answer often reveals a team that was not consulted and a manager who did not ask whether the customer wanted the extra at all.

A second, quieter trap is the answer that is really about recovering from a failure. The team missed a deadline, then worked hard to make it up, and the customer "ended up happier than before." That is a service-recovery story, and it can be a good one, but it is not the question asked. Delivering over and above expectations means the baseline was met and something was added on top, deliberately.

The strong answer does something different. It names the expectation in the customer's words, explains how the candidate found the gap between what was asked for and what would actually matter, and describes an addition that was small, targeted and visible. Underdog's demo is a useful mental model: the room is not a bigger model, it is the model's central claim made experienceable. The best "over and above" stories have that shape. The extra was the customer's own concern, answered before they had to voice it.

Applying STAR-T

Situation. Set the baseline expectation explicitly. "Our analytics customer had contracted for a nightly data export in a fixed schema, due by the end of the quarter. In discovery calls their analysts kept mentioning that they never trusted a new feed until they had checked it by hand for a few weeks." Name the stated expectation and the unstated worry in the same breath, because the gap between them is where the story lives.

Task. Say what you were accountable for, and keep it honest. "I owned delivery of the export on time and in spec. Nothing in the contract covered how the customer would validate it." The task should sound ordinary. If the baseline itself was heroic, you are telling a different story.

Action. This is where aim shows. "Rather than add fields or ship early, I had the team build a small reconciliation report that ran alongside the export and showed row counts and checksums against the customer's source system, so their analysts could see the feed was correct without the manual checking they dreaded. It cost us about four days of engineering, which I took from a lower-priority dashboard after agreeing the swap with my product partner." Note the three things a strong action paragraph contains: the choice of what to add, the reason it maps to the customer's actual concern, and the explicit cost with a named decision about who absorbed it.

Result. Measure the result in the customer's terms, not yours. "The customer signed off on the feed in the first week rather than the month they had planned for, and the reconciliation report was later written into the renewal as a standard deliverable." Then stop. A result paragraph that keeps adding adjectives is compensating for a thin one.

Trade-off. This question rewards naming a cost, so name it. "The dashboard we deferred slipped a sprint, and I had to explain to its sponsor why a feature for someone else's customer took priority. I would make the same call again, but I would have flagged the swap a week earlier." A trade-off stated plainly is what separates deliberate over-delivery from scope creep that happened to work out.

The follow-up that breaks weak answers

"How did you know that was what the customer wanted, rather than what you wanted to build?"

The heroics answer collapses here, because the honest reply is that nobody asked. The strong answer has evidence ready: a phrase from a discovery call, a support ticket pattern, a question the customer's team kept raising. If you cannot point to where the customer's real expectation came from, the interviewer will reasonably conclude that the extra was for you.

Score your answer against the director’s bar

Q: Tell me about a time when you delivered over and above customer expectations.

Ready when you are

Bank the story where the extra answered the customer's own doubt, and rehearse it until the cost comes out as clearly as the praise. Try it free →

Rehearse this in L8 Loop →