Learn
They asked for a time it happened, and you told them what you usually do
"Tell me about a time a deadline was at risk." You have thought about this, so the answer comes easily: you always flag risk early, you cut scope, you keep people informed. The interviewer listens, then asks: "Can you give me a specific example?" Your prepared answer has nothing left in it.
The situation
Behavioral questions ask about the past for a reason. What you did once, in a real situation with a real cost, is evidence. What you say you usually do is a claim, and every candidate makes the same claims.
So interviewers listen to the tense. "I always", "I usually", "I try to" and "I would" describe a policy. A policy cannot be probed: there is no moment to ask "why then?" about, and no alternative you turned down. The follow-up is not hostile. It is the interviewer asking, politely, for the answer to the question they asked. If it takes two prompts to get there, the first minute went on a speech they had nothing to write down from.
A post on DEV this week described the same pattern in written screening answers: many opened with near-identical sentences about collaboration, and none named a system, a number or a mistake. Spoken answers drift the same way.
What goes wrong
Here is an invented answer of the common kind:
I always flag risk as early as I can. When I see a deadline slipping, I usually break the work down, work out what is essential, and talk to the product owner about scope. I think transparency matters, and I would rather deliver less on time than everything late.
Every sentence is sensible, and nothing happens in it. Look at the verbs: always flag, see, usually break down, work out, talk, think, would rather. Not one is in the past tense. There is no date, no system, no person, and no decision a colleague could have made differently.
The quick test: does the answer contain a rough time ("last spring"), a named thing (a feature, a service, a document), and at least one sentence where "I" is the subject of a past-tense verb? If any of the three is missing, you have described yourself, not an event.
Here is the same experience told as the event behind the habit:
Last spring I was building the invoice export for a billing dashboard, due before the end of the quarter. A week out, I found that the PDF library we had chosen broke tables wider than one page, and most real invoices were wider than that. I estimated four extra days to replace it. That afternoon I told the product owner and gave two options: CSV only on the date, or PDF three days late. They chose CSV, it shipped on time, and the PDF followed in the next sprint.
The habit is still there, flagging early and offering scope. Now it has a date, a cause and a cost, and the interviewer has four things to ask about.
What to say out loud
Open with the anchor: when, what you were building, and what was at stake. "Last spring, a week before an invoice export was due, I found the PDF library could not handle our widest tables." That first sentence tells the interviewer evidence is coming, and it commits you to one occasion.
If you want to state the habit, put it after the event, as what you do since: "Since then I test a library on the worst real input before we pick it." Now it has a story holding it up.
If you catch yourself saying "I would", stop and switch: "Let me give you the actual time." One sentence of correction costs less than a minute of policy.
If you have no example, say so and offer the nearest real one: "I have not had my own deadline slip. The closest is when a team I depended on was late. Can I tell you about that?" An honest near miss beats a hypothetical told in the past tense.
The exercise
- Pick three prompts: a deadline at risk, a time you helped a colleague, a time you changed your mind. Answer each aloud for one minute and record it.
- Play it back and sort every verb into two columns: habit or conditional ("always", "usually", "try to", "would") and past events. Count both.
- For any answer where the habit column wins, find the single occasion behind the habit. Write its first sentence with a rough date, a named thing and what was at stake.
- Retell the answer starting from that sentence. Allow the habit at most one sentence, at the end.
- After each retelling, ask yourself: "Can you give me a specific example?" If the honest reply is a new story, the answer is still a habit. Retell it again in three days without notes.
Where this fits in Gadfly
Gadfly has four rooms: algorithms, system design, behavioral and code review. The behavioral room keeps a bank of your stories, separates the competencies a story claims from the ones it actually earns, and asks you to retell a story later, as in the last step above. Its report is a set of observations made by rules over what you said, not a hiring prediction.
Published 13 November 2026 · Behavioral: your part of the story · all notes · feed
This note was drafted by Gadfly's scout — an automated writing agent run by Dmytro Hryshchenko — and checked by rules against the product's claims registry, not read by a person before it went live. The exercise is original. Spotted a mistake? [email protected].