Should I be using estimated durations?

[JeremyE]

Almost all of my tasks are just ball park guesses as to how long they will take. For example, I have a task "Database Schema Design" that I guessed would take 31 hours and sure enough it only took 15 hours to complete. At this point to track my progress and I am going into the Task Sheet and entering "Actual Work" for that task. Then I go to the Gantt Chart/Work View and set the remaining work to zero which automatically sets the % complete to 100 %. But did I make the mistake of not setting ALL of my durations to estimated? I wasn't clear on when estimated durations should be used when I created the project plan, and I think that was a mistake. Could anyone give me some guidance? I can easily change all the durations to estimated, but I want to make sure that is what I should be doing.

Thank you so much!

 

[JulieD]

Hi Jeremy

just a thought -
2ed
doesn't mean an estimation of 2 days.
however 2d?
does
But when creating your project plan the aim is not to have lots of ? in the
duration column - these really are there to serve as an indication to you
that you need to research the tasks a bit more to get an idea of a realistic
timeframe for their completion.

The PERT Analysis toolbar is designed to help you determine durations, by
taking an optomistic estimation, a pesimistic duration & a realistic
duration popping them in a paper bag, giving it a good shake & coming up
with a duration that you can use in the project. The actual formula is
basically
task estimated duration = (an optomistic duration + a pesimistic duration +
4x realistic duration)/6

be a bit careful however when using this tool, if you don't fill in the
three columns in the pert entry sheet your duration column for that task
will be set back to 0 ... so if you're happy with the value in the duration
column, type it in all three of the other columns and only type different
things in the columns for those tasks where you're not really sure of what
the "right" estimation is.

Hope this helps
Cheers
JulieD

Almost all of my tasks are just ball park guesses as to how long they will

take. For example, I have a task "Database Schema Design" that I guessed
would take 31 hours and sure enough it only took 15 hours to complete. At
this point to track my progress and I am going into the Task Sheet and
entering "Actual Work" for that task. Then I go to the Gantt Chart/Work
View and set the remaining work to zero which automatically sets the %
complete to 100 %. But did I make the mistake of not setting ALL of my
durations to estimated? I wasn't clear on when estimated durations should
be used when I created the project plan, and I think that was a mistake.
Could anyone give me some guidance? I can easily change all the durations
to estimated, but I want to make sure that is what I should be doing.

 

[Steve House]

Marking a task duration "estimated" only sets a flag that turns the question
mark on - it really doesn't have any other effect. By the nature of project
planning all durations are estimates until you actually do the work. It may
be a real close to what it turns out to be but it's still an educated guess.
I think of the estimated setting as a marker I can use as a reminder to
myself to come back and revisit the estimate when perhaps I get better
information on what will be required to get the work done. It's a very rare
task that one can say "this will take exactly XX hours to complete, no more
and no less." You may *hope* your estimate is close to what it'll take and
plan a budget and schedule accordingly but most things are going to take
however long they take to finish the deliverable at its end.
--
Steve House [MVP]
MS Project Trainer/Consultant
Visit http://www.mvps.org/project/faqs.htm for the FAQs
..

Almost all of my tasks are just ball park guesses as to how long they will

take. For example, I have a task "Database Schema Design" that I guessed
would take 31 hours and sure enough it only took 15 hours to complete. At
this point to track my progress and I am going into the Task Sheet and
entering "Actual Work" for that task. Then I go to the Gantt Chart/Work
View and set the remaining work to zero which automatically sets the %
complete to 100 %. But did I make the mistake of not setting ALL of my
durations to estimated? I wasn't clear on when estimated durations should
be used when I created the project plan, and I think that was a mistake.
Could anyone give me some guidance? I can easily change all the durations
to estimated, but I want to make sure that is what I should be doing.

 

[Mark Durrenberger]

Jeremy,

To add to Steve's comments, if you don't have historical performance, then
all you have is an estimate.

To the best of your ability, try to get directly comparable experience when
making your estimate. And if you don't have the experience, find someone who
does or find someone who can find someone ...

And, as you run your project, keep track of actual results. Use the actual
results to improve your future projects.

FYI I teach a great program on estimating (he says modestly)...

Mark
--
_________________________________________________________
Mark Durrenberger, PMP
Principal, Oak Associates, Inc, www.oakinc.com
"Advancing the Theory and Practice of Project Management"
________________________________________________________

The nicest thing about NOT planning is that failure
comes as a complete surprise and is not preceded by
a period of worry and depression.

- Sir John Harvey-Jones

Almost all of my tasks are just ball park guesses as to how long they will

take. For example, I have a task "Database Schema Design" that I guessed
would take 31 hours and sure enough it only took 15 hours to complete. At
this point to track my progress and I am going into the Task Sheet and
entering "Actual Work" for that task. Then I go to the Gantt Chart/Work
View and set the remaining work to zero which automatically sets the %
complete to 100 %. But did I make the mistake of not setting ALL of my
durations to estimated? I wasn't clear on when estimated durations should
be used when I created the project plan, and I think that was a mistake.
Could anyone give me some guidance? I can easily change all the durations
to estimated, but I want to make sure that is what I should be doing.

 

[Steve House]

When you set the actual duration to xxx and the remaining duration to 0, the
scheduled duration for the task will change up or down to equal the actual.
If as in your case it's less than planned, this should in turn pull forward
the start of a linked successor task by the same amount the predecessor's
duration dropped. I say "should" because this is the intended behavior IF
the successor doesn't have any constraints on it which over-ride the normal
behavior. And just how do such constraints get there? Very often by
someone new to Project entering start and finish dates for the tasks when
they enter the task list instead of allowing Project to freely calculate
them. Manually entering a start date for a task sets a "Start No Earlier
Than" constraint and that instructs Project never to move the task earlier
in the schedule than that date even if it otherwise could be.


--
Steve House [MVP]
MS Project Trainer/Consultant
Visit http://www.mvps.org/project/faqs.htm for the FAQs

Thanks for your thoughts on this. I will just continue adding "actual

work" instead of mesing with estimated durations, etc. But, my last
question (ha!) is how I should be changing the start dates of tasks. For
example, this 25 hour database planning task only took 15 hours, meaning I
can start the next task more than a day ahead of schedule. Will I have to
manually tell project I want to start the next task early, or will it see
that I completed the first task early (when remaining work is 0) and
automatically bump up the start date for the next task a day early? I know
I should just test it myself, but I feel more comfortable getting a second
opinion. Thanks!!

 

[JeremyE]

Yep, Project is working the way it should be, according to your post. Thanks.
--

Jeremy

When you set the actual duration to xxx and the remaining duration to 0, the
scheduled duration for the task will change up or down to equal the actual.
If as in your case it's less than planned, this should in turn pull forward
the start of a linked successor task by the same amount the predecessor's
duration dropped. I say "should" because this is the intended behavior IF
the successor doesn't have any constraints on it which over-ride the normal
behavior. And just how do such constraints get there? Very often by
someone new to Project entering start and finish dates for the tasks when
they enter the task list instead of allowing Project to freely calculate
them. Manually entering a start date for a task sets a "Start No Earlier
Than" constraint and that instructs Project never to move the task earlier
in the schedule than that date even if it otherwise could be.


--
Steve House [MVP]
MS Project Trainer/Consultant
Visit http://www.mvps.org/project/faqs.htm for the FAQs

Thanks for your thoughts on this. I will just continue adding "actual

work" instead of mesing with estimated durations, etc. But, my last
question (ha!) is how I should be changing the start dates of tasks. For
example, this 25 hour database planning task only took 15 hours, meaning I
can start the next task more than a day ahead of schedule. Will I have to
manually tell project I want to start the next task early, or will it see
that I completed the first task early (when remaining work is 0) and
automatically bump up the start date for the next task a day early? I know
I should just test it myself, but I feel more comfortable getting a second
opinion. Thanks!!

 

[Steve House]

Glad it helped - the object lesson is one should never, ever, manually input
the dates for the tasks unless there is a specific reason to set a
constraint - ie, maybe a critical part just won't be available from a vendor
before a certain date, for example, so it doesn't do any good to schedule
the task that uses it to start earlier than that date cause it just won't
happen any earlier regardless of what your plan says or how badly you want
it. To put it succinctly, Project's job is to tell you what dates you
should tell the resources to work on what tasks, NOT to simply draw fancy
pictures of dates that have already determined or that your resources have
told you are convenient for them <s>. Put bluntly, the job of the Project
Manager is tell the resources where to go when and what to do when they get
there and MS Project's job is to help the PM figure out what to tell them
<wide grin>.

--
Steve House [MVP]
MS Project Trainer/Consultant
Visit http://www.mvps.org/project/faqs.htm for the FAQs

 

[Mark Durrenberger]

Steve,
you have to admit though, project does make pretty pictures

Mark

--
_________________________________________________________
Mark Durrenberger, PMP
Principal, Oak Associates, Inc, www.oakinc.com
"Advancing the Theory and Practice of Project Management"
________________________________________________________

The nicest thing about NOT planning is that failure
comes as a complete surprise and is not preceded by
a period of worry and depression.

- Sir John Harvey-Jones