Tell me about a time you had to tell a stakeholder that work was going to be late.
answer
- the moment you knew it would slip
- how fast you said it out loud
- what you offered with the bad news
- the new date and whether you held it
- what you changed to see it sooner
basics
~20 sTests whether bad news travels fast and arrives with a plan. Answer with the moment you knew, how quickly you said it, what you offered alongside the slip, and the revised date you then held.
how to answer
6 beats- what you had committed to and to whomOne or two sentences: the deliverable, the date, and who was depending on it for what. Keep the setup tight — the interviewer is waiting for the conversation, not the project brief.
- the signal that told you it would slipName the concrete moment of discovery and, if you can, the measurement behind it. This beat quietly answers the follow-up about whether you could have known sooner, so make it specific rather than saying the work turned out harder.
- how and when you delivered the newsSay how many hours or days passed between knowing and telling, and why you chose that channel — a live conversation for a costly slip, a message for a small one. Speed and directness are the primary signal in this whole answer.
- what you brought with itNever just the bad news. Describe the partial delivery, workaround, or reduced scope you offered so the stakeholder kept something on the original date, and the revised estimate you could defend.
- the revised date and whether you held itClose the loop with the actual outcome. Result and reflection together should run around a quarter of your airtime, and a story that never says what finally happened leaves the interviewer assuming the worst.
- the early warning you built afterwardsFinish with the practice that came out of it — measuring a slice before quoting, a reconciliation week, a checkpoint at a third of the estimate. Say it in one sentence and stop.
your answer
5 story prompts- Pick a slip you communicated yourself, not one your manager announced on your behalf.
- Write down the date you knew and the date you told them; the gap is the story.
- Name what you offered alongside the news: partial scope, a workaround, or a way to verify progress.
- Have the revised date and whether you actually held it ready before you rehearse.
- This can be the same project as your estimation story, told from the conversation rather than the plan.
draft and rehearse your own answer in a learn session
go deeper
This prompt probes ownership and communication under pressure, and specifically the instinct to make bad news early rather than comfortable. Interviewers want evidence that you separate the delay from your ego, that you bring options and a revised date rather than an apology, and that you protect the stakeholder's ability to plan around you.
This was in my first year on a data team at a small company. I owned one piece of a new signup-source dataset, and I had said the historical load would be ready for a Wednesday review with our head of marketing. On Monday I ran it against a tenth of the history to sanity-check myself, and the result was bad: extrapolated out, the backfill runtime came to roughly twenty-six hours, not the overnight job I had assumed, and I had one machine to run it on. My first instinct was to try to make it work quietly and hope. I did not, mostly because my mentor had told me a surprise on the day is far worse than bad news two days early. So I messaged the marketing lead that same afternoon. I told her plainly that Wednesday was not going to happen, and I brought two things with it: a specific new estimate of Thursday afternoon, and an offer to load the most recent sixty days in time for Wednesday so she could still validate the field definitions in the review. She took the sixty days, and the review went ahead. The full load finished Thursday at two. Since then I always run a slice of anything long before I put a date on it.
For a junior answer the signal is entirely in the speed and the accompanying offer: told the same afternoon, with a defensible new estimate and something the stakeholder could still use on the original date. Naming the temptation to stay quiet reads as honest self-awareness. It would weaken if the news arrived on Wednesday, however well phrased.
I owned migrating our subscription metrics onto new warehouse models, and the date had been shared with finance because their close process depended on it. Three weeks in I found that the old system had been silently deduplicating trial records in a way nobody had documented, and reproducing that faithfully was real work. What made me escalate immediately was the shape of it: this was not a slower version of my plan, it was a different problem, and those never recover quietly. I took it to the finance stakeholder and my product manager together, in a fifteen-minute meeting rather than a message, and I brought three things. The size of the discrepancy — about four hundredths of a percent of reported revenue drifting between the two systems. A revised date eleven days out. And a proposal that we run both systems in parallel for one close, so finance could reconcile rather than take my word for it. I had checked first that the extra nightly job cost us about four hours ten of backfill runtime we could afford. The parallel close bought the goodwill, because it gave them a way to check me instead of waiting on me. We landed the revised date and the two sets of numbers matched inside the tolerance they set. Now I spend a week reconciling against the old system before quoting any migration date.
The middle-level signal is choosing a live meeting with both stakeholders, arriving with cause, a revised date and a verification mechanism, and having already checked the cost of what was offered. The parallel-run proposal is the move that converts bad news into trust. Without the new date being held, this would read as competent communication with an unproven outcome.
This is squarely your question, and owning one task well is the whole bar. Show that you noticed the slip yourself, said it within hours rather than days, and arrived with an estimate and something useful you could still deliver on the original date.
You are expected to manage a stakeholder, not just inform one. Show that you chose the channel deliberately, brought the cause and a revised date together, and offered a way for them to verify progress instead of asking them to wait and trust you.
The bar is that nobody was surprised, because your team's delivery signals were visible before the slip. Show that you took the news to the stakeholder rather than letting an engineer carry it alone, renegotiated scope, and fixed the forecasting gap that hid it.
Speak to how slips surface across an organisation. The signal is a reporting rhythm where dates degrade publicly and early, so bad news is routine rather than an event — and evidence that you removed the incentive to hide it.
saying these in an interview costs you the question
- Waiting until the deadline itself to raise a slip you saw coming
- Delivering the news with no revised date and no partial scope
- Blaming another team or the original estimate rather than owning the communication
- Burying the slip in a status document instead of telling the person directly
- Promising a recovery date with no basis and missing that one too
- Framing the stakeholder's disappointment as unreasonable
- How early could you have known?Answer honestly and specifically. Almost every slip is knowable earlier than it is spoken, and interviewers want to see you can identify the missing signal — an unvalidated assumption, an untested slice, a dependency you never confirmed. Then say what you now do to surface it, which turns a confession into evidence of learning.
- How did they react, and what did you do with that?Do not sand off the reaction. If they were frustrated, say so and describe what you did next: acknowledged the cost to them, avoided defending your estimate, moved the conversation to options. Candidates who report only calm acceptance sound like they are narrating an ideal rather than remembering an event.
- Did you hit the revised date?Say plainly whether you did. If you did, mention what buffer or scope cut made it credible. If you did not, own the second miss and what it taught you about padding your own recovery estimates. A slip followed by a second silent slip is the worst version, so show that the second conversation happened even earlier than the first.