Do you estimate in Days or in Hours?

Some of you might say, neither. So let us just assume I am in a situation where I am required to use some unit of time to measure my estimates in. Then which is the best unit to pick. Should we estimate in Days or in Hours?

In a lot of teams, I have noticed that people use hours to estimate work. And then I have also seen they come up with a multiplier which reflects productive hours in a day. This can vary from 3 hours to 6 hours per day based on a team’s situation. But then I notice an interesting thing. If you closely look at the estimates, you’ll see a pattern which can easily tell you the multiplier used by the team. If the team thought 4 hours per day was the productive time their members have, you’ll typically see the estimates in multiples of 4. If they used 6 hours as productive hours multiplier, most of the estimates are in multiples of 6.

So that means, we originally think in terms of days, and then convert them to hours based on whatever multiplier we use. The the question is why not just use days as the estimation unit. For smaller tasks we can use .25 days, .5 days etc.

I see the following advantages using Days as estimation unit:

• I find it easier to say how much I’ll get done in a day rather than how much I’ll do in certain hours.
• If different teams in the organization have different multipliers, estimating in days will remove the inconsistency in their data. 200 hours for one team might not be same as 200 hours of another team, but 25 days of one team will be the same as 25 days of another team.
• Tasks estimated in days give a team member better indication of when the work will be done as compared to if the estimates are in hours. e.g. if I see that I have 15 days of work remaining I have a better idea of when I’ll be done, as compared to if I said I have 90 hours of work remaining.