using MS Project for tasks tracking- not hours/effort or work tracking

[Lisa@IMC]

Hi,

I am new to MS project and have taken basic classes and need help with an odd situation.

I am managing a software implemetation project- only the completion of tasks are of concern, I don't track how many hours or amount of effort. Additionally I am not concerned with costs or leveling.

An example of my typical tasks: Training has to be scheduled ( I don't care how long it takes only that it gets done by x date); the client has been contacted with specific information etc. ( only that it gets done by x date)

I have 2 questions:

1. How to schedule tasks that won't take 20 days to complete but are due in 20 days. I am using the following Predecessor type: 20FS+20d, with a duration of 1 day. Is this the best way to utilize Project?

2. I have a task that can not be completed due to clients hardware issues, do I reschedule it using the reschedule function or change the predecessor (probably a bad idea I know)

I was originally using MS Excel but management has decreed I start using MS Project.

Thanks for any input!

Lisa@IMC

I am in a quandry what to say. You say that you are only interested
"the completion of tasks. Yet you ask in your questions about "how to
schedule tasks", and how to "reschedule" tasks. Further, you indicate
that "management has decreed" that you use Microsoft Project.

My suggestion would be to do the following:

: go back to Management and find out their expectations. It's my hunch
that they expect more of you than simply tracking the completion of
tasks. I suspect they are asking you to plan and schedule the project.
If all you are doing is tracking completion of tasks ... maybe just use
a piece of paper or Excel if you wish to be fancy. Find out from
management who defines the task completion dates.

If they have in mind a larger role for you:

: go to the person who now has the plan (I assume it is not you?) and
find out from them if they are using Project or not. If there is nobody
else (and I suspect that the case) then following on from your
conversation with Management, work with the Project Manager or whomever
created or otherwise is responsible for the plan.

: get some Microsoft Project training. If that not possible, at least
buy one of the books that are now easily available and read and experiment.

: use Project to define the plan. You'll need to care about durations,
logical sequence, external constraints ... a whole bunch of things.

Cool. Now we have something to work with. Will cogitate and get back to
you soon.

[Lisa@IMC]

Prior to my managing the projects there wasn't anyone that tracked it. This is a new venture for the company (we were recently purchased and thus my role was created). I am responsible for defining the plan and completion dates so I am the Project Manager. I have taken 2 MS project classes and understand the concepts.

We sell an off the shelf product that does not require software development, we do not charge for my or anyone else's time and are not worried about costs or budgets. My role is to ensure that stuff gets done when I determine it should.

The CEO has told me that he wants me to start using project to schedule and track the implementations, and some of our RFP's require that we provide a plan in project.

The role that new management has in mind is that I ensure that the following things occur:

• communication with the Client regarding the process & require responsibilities on their part
• installation of the purchased software
• scheduling and completion of training
• configuration of software interfaces
• scheduling of software "Go Live" with the client
• tracking of data conversion projects. The start and end dates for these are determined by me and the programmer.
• determination of risk and communication to stakeholders

Some of these items have to occur in a specific order and within certain number of days prior to "go live"

When I referenced rescheduling a task I was referring to an issue that has just occurred:

A task was slated to finish on 7/7/09. My tech just informed me that the client does not have the required hardware to complete the task and is unsure when they will have said hardware. I am wondering do I leave it as a slipping task or reschedule it for completion once the HW is in place.

So I guess in summary: I am required to use Ms Project, ( which I like using). I am looking for feedback as to best practices for my situation.

Thank you so much for your input, it is an odd situation. I like the idea of milestones, here is a pared down list of the tasks that must occur before a client can go live with our software:

• Sales Hand off to project manager
• Internal Implementation meeting
• Contact vendor for data extraction
• Advise vendor of Database Training Date
• Assign Trainer as Training Liaison
• Mobile Technician Assigned
  
• Software Installation on client's server
• Configure workstations (by client)
• Database code entry
  
• Tech to Contact Client with Mobile information
• Verify Mobile HW is in place
• State Notified of Vendor Switch by Client
• Mobile installation appointment Scheduled
• Mobile Switch Configuration
  
• Interfaces configured
  
• Training Scheduled
• Training dates completed
  
• Confirm and Notify staff of go live date
• Notify Data Extraction Vendor of go live date
• Go Live prep meeting
• Configure system for go live
• Go Live

Many of these tasks are performed by the client and many by my company. Some have to take place sequentially and some do not i.e. interfaces configured can occur at anytime prior to the go live date, unless the client opts to wait whereas the mobile configuration tasks have to occur in order.

So would I make milestones of all of these (I am thinking not) and do I continue to link tasks that need to occur x number of days before go live? I had thought about scheduling from the finish, but if that date changes what are the ramification (they often do change due to client issues).

Maybe I will try the schedule from finish date as test.

If your plan had as a pre-requisite the task named "Hardware in Place",
the slippage (non-completion) that task would push out the planned start
dates for subsequent tasks that depend on that.

I amy be wrong. At this stage I would normally roll up sleeves and have
a look a at the MPP file. I haven't done that. I can only guess based
on what you write.

I have a hunch that your MPP file in Project for this project is
constructed from the perspective of just recording the dates for
completion of tasks. I suspect you have a list of those tasks and have
entered start and/or finish dates, along with some durations which when
printed as a Gantt makes it look like you have a plan. Then when things
don't happen as planned, e.g. the hardware problem mentioned, you have
to rejiggle the whole plan. this isn't the way to use Project. Entering
dates into start/finish is never the right thing to do--unfortunately,
nothing in Project stops you from doing that. Project will only get in
the way and be a pretty much useless to you when used this way.

I can't write in a few sentences here how to make Project work right. I
do know that the right advice is sometimes elusive. You need to build
a model of the project with tasks that have successors and predecessors
that do not define start/end dates. Let Project compute those dates for
you. Then as things change, Project will reschedule for you based on
the model. You can change the model if that required, but in general
the model (successor/predecessor) relationships are not changed unless
the execution strategy changes. I think you do need to think about and
define budgets not from the perspective of cost, but from the
perspective of estimates of work days for tasks to check for
reasonableness of the plan. Things get done when they can be done--not
necessarily "when you determine it should be done" in absence of a
reasonable plan. A reasonable plan has to reflect the logic, the work,
the resource availability, external and internal constraints, and a
whole lot of things. And then even then the best plans are at risks due
to unanticipated and unintended "gotchas".

See , and in particular
http://project.mvps.org/mike's_tutorials.htm. The resource I like the
*best* is
http://www.amazon.co.uk/Dynamic-Sche...&creative=8898

Let's give you a solution that is easy to acomplish your 'odd situation'..

Work is scheduled over time using a duration to decsribe that amount of time.

You aren't interesd in plotting time, just the finish, right?

Starting a task - like training - or finishing training is an event; a point
in time without a duration.

This is handled in your case by a milestone; a zero duraton point in time
that you can use and still save a baselne before you start tracking the
actual dates and compare when they actually finished training, for instance,
against when it was originally scheduled.

That gets it done, right?

I like that idea as a great and easy way use the tool to get visibility
of the status of the program against the baseline. It doesn't use
Project to help develop/compute the plan but I think it gets close to
what Lisa is saying she wants. Good catch.

I liked your approach too - after I pressed 'send' it occured to
me I should have mentioned that your advise was well placed to handle the
management team for communication and train them on the benifits of
exploiting the power of MS Project - but, it sounds like you would agree
thats a tough one to engage new bosses. In any case - thanks again buddy!

I call it "Planning by PowerPoint".

(Works for some... )

1) Use the Deadline feature.

2) If the task has started, then insert a split, with the task resuming
on the date you expect the hardware issues to be resolved. You can also
have project do this somewhat automatically for you: Tools > Tracking >
Update Project > Reschedule uncompleted work to start after > for
selected tasks. for this to work, in Tools > Options > Schedule, check
the box to "Split in-progress tasks."

1. Do not schedule from finish date. All that gets you is a "as late
as possible" schedule which by definition is at risk of not being
reasonable.

2. Surely you now can "wire-up" these tasks in some sort of logical
order. You know which are predecessors and successors? That will give
you a good schedule. Build it as a "dynamic" schedule. See
http://www.amazon.co.uk/Dynamic-Sche...&creative=8898

3. Milestones (tasks with zero duration) will be when groups of these
tasks are complete. The predecessors of milestones would be the tasks
that need to done to declare the milestone complete. If you know when
you expect the milestone to be achieved (based on your reasonable plan
that you baselined!), then set the deadline field to be that date. Let
Project help you. It will help you tell if future milestones are at
risk of slipping past the deadline.

4. Because of the nature of this work, I would be doing a Monte Carlo
simulation (with add-in tools) of the plan before committing to deadline
dates. That only possible if you can do No. 2 above. Since it's someone
else's money, I would assume that knowing the risks is a high priority.

The key is to build a "model" of the project in Project.

Coincidently I concluded that I could use your example in a session with
someone this afternoon to describe how to use Project at a "higher"
level for achieving

: overarching view of the programme, but not as a "scheduler" for
purposes of setting up a reasonable plan which includes the logical
progression, and sets up the key milestones and deadline dates

: become the main communication vehicle for the project team and
stakeholders about the plan, the current status, and forecast. This is
the main purpose and used as an example for what we will do for their
project.

Thus I expressed it in a partially-complete MPP file that we will flesh
out for our own purposes--going through how to do it and imagining a plan.

I'm happy to share with you this MPP file if you ping me at my email
address (figure out from the email I show in this posting) with your
email address. Easier to show you this way than it is to explain in
words here in this forum.