How would you handle a task you don't know how to do?
Say you'd first clarify what done looks like and when it's due, then spend a short, fixed time finding out yourself from documents, past examples and tools. Next, ask the right person a specific question that shows what you've tried, and check in early with a first draft rather than at the deadline. Close with a short real example of doing this.
The balance they're listening for
This situational question has two wrong answers, at opposite ends:
- Too helpless: "I'd ask my manager how to do it." You'll be asking them about everything.
- Too proud: "I'd figure it out myself." You might spend three days going the wrong way, or get it wrong without telling anyone.
The right answer sits between: try first, for a set time, then ask a good question, then check in early. That shows you're independent and safe to give work to.
A four-step answer
- Clarify the goal. "First I'd make sure I understand what a good result looks like and when it's needed." Ask this at the moment you get the task. It's not a sign of weakness.
- Look for yourself, for a set time. Past examples, shared drives, templates, documentation, the tool's help pages, a colleague's old version of the same work. Say how long: "I'd give myself an hour or two."
- Ask a specific question. Go to the person most likely to know, show what you've already found, and ask about the one thing you're stuck on.
- Check in early. Share a rough first version well before the deadline, so any wrong turn is cheap to fix.
Good questions versus helpless ones
The difference between sounding capable and sounding lost is mostly how you ask:
| Helpless | Capable |
|---|---|
| "How do I do this?" | "I've found last year's version and I've updated sections 1 to 3. For section 4, should I use this year's figures or the forecast?" |
| "I don't understand the system." | "I can run the report, but the totals don't match finance's. Is there a filter I should be applying?" |
| "Can you show me?" | "Could I watch you do the first one for ten minutes, then do the next one myself while you check?" |
| Asking on deadline day | Asking on day one, with a draft by day two |
Put one of these "capable" lines into your answer. It makes the approach concrete.
Example answers
These are examples with made-up details. Use a short real story from your own experience if you have one.
Example answer for an office or admin role
"First I'd check exactly what's needed and by when, so I'm not guessing at the goal. Then I'd look for a past example, because most tasks have been done before in some form. I'd give myself a set time, say an hour, to work it out from that and any guides. If I was still stuck, I'd go to whoever knows it best with a specific question, like 'I've done these parts, I'm stuck on this one.' And I'd share a rough version early. In my last job I was asked to set up a supplier contract in a system I'd never used. I found an old contract as a template, did most of it in a morning, and asked our finance lead one question about payment terms. It was approved first time."
Example answer for a technical role
"I'd start by making sure I understand the requirement and how it'll be tested. Then I'd read the existing code or documentation around it and try a small spike, maybe half a day. If I'm still stuck, I'd write up what I've tried and ask in the team channel or pair with someone. I'd rather spend ten minutes of a senior's time than two days going the wrong way. And I'd open a draft early so reviewers can catch problems before I've built too much."
Example answer for a graduate or entry-level role
"I'd ask questions upfront about what's expected, then try to work it out from any examples or guides. As someone new, I know I'll have to ask for help, so I'd try to ask good questions: show what I've done, then ask about the part I'm stuck on. In my internship I was asked to build a pivot table, which I'd never done. I watched a short tutorial, built a first version, and asked my supervisor to check it before I sent it on. She fixed one setting and the rest was right."
Mistakes to avoid
| Mistake | Why it hurts | Try instead |
|---|---|---|
| "I'd ask my manager straight away" | Sounds dependent | Try first, then ask a specific question |
| "I'd never ask for help" | Sounds risky and proud | Show when and how you'd ask |
| "I'd Google it" with nothing more | Vague | Name the sources: past work, docs, a colleague |
| No time limit on struggling alone | Can mean days lost | Say how long you'd try before asking |
| Saying you'd pretend you know | Damages trust | Say you'd be honest about what's new to you |
Practice it out loud
Interviewers often follow up with a real task from the job: "So how would you approach our quarterly audit?" Practice answering a made-up version out loud, using your four steps, in under a minute. If you want a partner who gives you that kind of follow-up based on what you said, Speaking Coach's mock interviews do that. Also see how to answer a question you don't know for the moment in the interview itself.
Questions people ask
Is it okay to say I'd ask for help?
Yes. Asking for help is expected. What matters is that you try first and ask a clear question, not a general one.
How long should I try alone before asking?
It depends on the deadline. A useful rule to mention: a small fraction of the time you have. For a task due in a week, an hour or two; for one due tomorrow, ask sooner.
Should I give a real example?
If you have one, yes. A 20-second story after your steps makes the answer much stronger.
What if the task is something I really can't do, like a skill I don't have?
Be honest. Say what you'd do to learn it, and offer to pair with someone who has the skill while you get up to speed.
How is this different from "Tell me about a time you learned something quickly"?
This question asks what you would do; that one asks for a past story. One clear method covers both. See learning something quickly.