When you’re given a story, task, or vague statement and asked for an estimate you need to think about how “complete” the thing you’re estimating is. If it’s a small task that matches something you’ve done a lot before there’s probably not a lot of unknowns so you can estimate based on what’s there. If it’s a larger task, or something not well defined, then any estimate based on the current description is going to be wrong.

You can think about estimating how complete you think the description is, and how likely it is to change. You will likely need to know something about the wider context to be able to do this. If it’s pretty well defined but for a customer you know changes their mind a lot then add some time to the estimate for that. If it’s not well defined it’s more likely to get bigger than smaller in scope and so your estimate should take that into account.

If ignore the unknowns and try to minimise your estimate to make whoever is asking happy then you’re setting yourself up for failure. If they’re not happy with the estimate then that’s the time to start the conversation on scope and resources.

Do you think about how complete a description is when you’re asked to estimate it?