I have long wanted to have a down-and-dirty tool to help me improve my
everyday estimating. I'm not talking about rocket science, with five
decimal places and standard deviations, just something like this:

My boss asks me how long it will take me to do X, where X is a task
that (let's say) is a work day or less. Maybe the task is to prepare a
project work estimate to be included in a proposal. :-)

Anyhow the tool I imagine would support me through this loop:

1) I estimate how long it will take me
2) I keep rough track of how long it takes me and enter it into the
tool
3) There is a panel that shows the original estimate, the actual time,
and a comments place for me to write down how I could improve the next
similar estimate
4) There is a report or something that allows me to see my estimates,
what the variances were, and how I thought I could improve them. Maybe
it could be sorted chronologically, and by variance.

That's it.

I don't want to get sucked into a discussion of how MLO (or any other
similar tool) should be turned into a time and billing accounting
tool. That's overkill for this.

The way that I use MLO, this facility would be useful only one layer
up from the bottom, I think. That is, my experience has been that
there is so much variation and redefinition at the bottom level of
tasks that one can't track an item well or at all. Let's use the
example of preparing a work estimate for a change to be made to some
functioning software. At least in my line work environment things are
not routine but rather exploratory, and the bottom-level tasks keep
metamorphising -- they change their definition or balloon or deflate
constantly. But one level up (roughly), at "prepare and submit work
estimate to support proposal Z," there is continuity and closure of a
kind.

For the purposes of this discussion I'd like to define "estimate" as
the duration that you think the actual is equally likely to be above
or below. That is, it is not one of the several varieties of
"estimate" that exist in the real world, it's a judgment that from
what I know right now, the actual number will be equally likely to be
above or below the "estimate" I am giving. Not a ceiling, not a floor,
not a promise to someone, just a judgment of simple reality. And the
idea is to be able to do this better (with less variance) over time,
with the help of this tool.

I'm happy with the existing, loose tools that MLO uses for entering
effort and time estimates. And I don't want to complicate MLO. But I'm
trying to envision some sort of support for an estimate improvement
loop, so I've posted this message so people can talk about it.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"MyLifeOrganized" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/myLifeOrganized?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to