How to Estimate How Long a Task Will Take
Most of us are bad at estimating time. We remember our best runs, not our average ones, and we estimate as if nothing will go wrong. Here is a method that produces estimates you can actually plan around.
Why your estimates are usually wrong
The planning fallacy is well documented: people consistently underestimate how long their own tasks will take, even when they have done similar tasks before. The trick is not to try harder — it is to estimate differently.
Estimate ranges, not points
Instead of "this will take 30 minutes", say "this will take 20 to 60 minutes." A range forces you to think about both the best and worst case and gives you something honest to plan against.
Write down your assumptions
A 30-minute task assumes you know the tool, are not interrupted, and have everything you need. List those assumptions explicitly — when one breaks, you know which estimate to revisit.
Break it down first if you are wildly off
If your gut says "an hour" but the worst case feels like "all afternoon", the task is too vague to estimate. Break it into smaller chunks, estimate each, and add them up.
Track estimate vs. actual
Write down how long you thought it would take, then how long it really took. After ten tasks you will see your bias clearly — usually about 1.5x your initial guess — and you can compensate.
Pad for the invisible work
Most estimates only cover the obvious middle of a task and forget the edges: finding the file, remembering where you left off, the email that interrupts you, the tidy-up afterwards. This setup and switching cost is real and it adds up. A simple fix is to estimate the core work honestly, then add a fixed buffer — often 20 to 50 percent — for the wrapper around it.
Try it with Mibbi Estimate
Guess how long things might take.
Open Mibbi EstimateFAQ
- How wide should the range be?
- A 2x-3x ratio between min and max is normal for unfamiliar work. For tasks you have done many times, a tighter range is fine.
- Should I tell other people the range or a single number?
- Share the range, or at least the top of it. If someone needs one number to plan around, give them the higher end — under-promising and finishing early builds trust, while a tight single estimate that slips does the opposite. Honesty about uncertainty is a feature, not a weakness.
Related guides
- How to Stop Underestimating How Long Tasks Take The planning fallacy is universal — and beatable. Here is how to stop underestimating and size tasks the way they actually unfold, not how you hope.
- How to Estimate a Creative Task Honestly Creative tasks resist estimation — but not as much as we pretend. Here is the method that produces useful ranges.
- How to Quote Time to a Client Honestly Honest time quotes protect both you and the client. Here is how to give one without underbidding yourself or padding the number just to feel safe.