Apart from the social issues there‘s a nice cognitive corrective described in „heretics guide to best practises“. 1. optionally divide your project in arbitrar

Apart from the social issues there‘s a nice cognitive corrective described in „heretics guide to best practises“. 1. optionally divide your project in arbitrary smaller tasks 2. for each subtask pick a range (min and max estimate) so that you‘re sure the actual time lies in the range 90% of the time. 3. imagine the following decision: you can draw one out of 9 black and 1 red marble. If you pick the red you get 10.000$. That‘s a chance of 90%. OR: you can choose your estimate and get 10.000$ if it‘s correct. If you think about this decision and rather pick the marble-game, then your estimate is too narrow and most likely not correct. If you pick your estimate over the marble game, it‘s too wide, because you‘re more than 90% sure about it. Adjust your estimate until the decision doesn‘t matter anymore and both the game and your estimate seem like equal chances to you. 4. my own addendum for people who are afraid of asking too much: double the result to take the stress of yourself.