As mud.
The "Work" value is an estimate, a planning placeholder. Once the project begins all you have to worry about is actual work and
remaining work
on a task. If you need to look backwards and find out where the plan was incorrect then take a baseline. Or rather than load the
system with baseline data
just use a custom number field with your original "Work Estimate" which will remain non calculated.
It goes like this task A is planned as having 30 hours WORK
once task A begins it has a certain amount of ACTUAL WORK added to it in a tracking operation
If the ACTUAL work is recorded (the first and most logical step) then the REMAINING WORK is calculated
REMAINING WORK = WORK - ACTUAL WORK
But wait this is real life here, Once the task is in the hands of the resource and they are in full control of it's plight. The
resource knows how best to bring their
skills to bear on it and how many times they will be interupted while trying to get it done and have looked under the hood of the
task and are better qualified
to make a judgement on how much of their ACTUAL WORK is required to accomplish the task. So f...k the plan. Ask your resource to
tell you how much
work they have done as of right now and how much work they have REMAINING. ENTER those two values. in that ORDER.
When the task is at hand you now have the world's best expert on task estimation at your disposal, the resource itself. Use that to
your advantage and
you will be setting REMAINING WORK to ZERO much more frequently and accurately this way.
So in answer to this
my manager thinks we should adjust the 'work' to 10 hours to match the 'actual work'.
Don't enter WORK, once the plan has been put in motion let it be calculated by Project by sticking to a simple disciplined rule.
enter ACTUAL WORK then REMAINING WORK. As often as possible.
As an aside scenerio,
If you resource happened to see a task that was a full week and he knew he would have it done in the next 3 days. Wouldn't you want
to know after 3 days
that REMAINING WORK is zero. Now that may be an extreme case but multiply even a portion of that ESTIMATE vs ACTUAL variance by a
large number
of projects each with a large number of tasks and you can see the power of having you resources communicate their status as
frequently as possible.
cue the chorus who think they can see into the future better than the rest of us.