Mike presented a summary of results from a presentation by Magne Jørgensen and Stein Grimstad: How to avoid impact from irrelevant and misleading information on your estimates. Three groups are given the task of estimating the same project. The control group comes up with an estimate of 456 hours.
The second group is told that the customer thinks the task should take 500 hours. They are also told that the customer has little basis for this, and they should disregard the customer's expectations in their estimate. In effect, this customer input should be irrelevant to the estimate. This second group comes up with an estimate of 555 hours.
The third group is told that the customer expects the task to take 50 hours, but again to disregard the customer's expectation in the estimate. This group estimates 99 hours.
Think about this the next time you ask your team for an estimate. I'm sure we can all hear ourselves saying, "I think it should take about a month" or "I'd like to know if we can get it done by the end of the year."