|
Question: Mary Cahill
I'm
looking for ideas about quality assurance for projects and project
managers. I am tasked with developing a 'Quality Plan' template for IT
projects and am trying to determine how to measure the effectiveness of the
PM process. Quality objectives for the 'solution' are fairly easy, as are
cost and schedule objectives. But what about the actual process? Most
Quality Plans I've come across are based on the ISO 9000 elements, and I am
familiar with them. This organization is not ISO certified and I am not
looking for objectives based on test faults, faults in production, etc.
If a
PM must have a Quality Plan for their project, they must tailor it to suit
the project. Other than simply stating the obvious, i.e. the project will
be managed according to the organizational PM methodology, how else do you
define 'quality' process for a project?
Plan
and define the deliverables (Scope, Requirements, Design, etc), number and
type of reviews, then check to see if they followed their plan? Monitor the
number of 'changes' to scope or requirements?
Or do
you go back to the old faithful checklist, e.g. is there evidence of WBS;
baselines for cost and schedule; major milestones; key risks identified;
assumptions and constraints; approvals verifying acceptance of project
products; etc.
Am I
trying to make this too difficult? Any advice would be greatly appreciated.
Answer:
Adam Pearson
I
admire your goals: "developing a 'Quality Plan' template for IT projects and
[] trying to determine how to measure the effectiveness of the PM process".
I also note your focus on the "quality" of "the process" - I interpret that
you are satisfied about managing the quality of the "result(s)".
Is it
as simple as saying that "quality" is "meeting expectations"? I am sure
that you and others will have a stake in the "quality of the process" - is
it meeting your expectations, which must mean - "does the process appear to
be going to lead to the results you want?". If this is the case, then what
is needed is a set of techniques for monitoring this "success (quality) of
the process". At first glance, this would seem to suggest a set of metrical
measurements, based on your forecast development of a number of success
factors.
But I
assume from your post that you are seeking more than these metrics - maybe
you are looking for ways of confirming "owner" and "stakeholder"
satisfaction with the process - is it giving them confidence that the
results will be achieved, are they informed enough to know, are they
contributing where they should? What you need is a means of achieving owner
and stakeholder commitment right up front.
Answer: Mary Cahill
I
appreciate your thoughts on this subject. Someone else recommended using
the Competency Standards and PMBOK Guide. The issue with basing a Quality
Plan on the competency standards is that you are focusing on the individual
(PM), and while this can be good, it detracts from the 'purpose' of the
Quality Plan. Yes, we want to ensure the PM is 'doing the right thing' but
the end result is the solution delivered to the business. Then the age old
dilemma presents itself - time vs. cost vs. quality, and quality invariably
loses out.
I
suppose I should get the PM community to agree to the purpose/intent of the
quality plan. The boss wants this deliverable, I don't want to see another
piece of documentation issued that is 'ignored', and the PMs don't want any
extra work that doesn't deliver value to them or their customer.
My
other 'assignment' was to get the buy-in of the PM community, but I could
only get actual feedback on the draft plan from 2 of them. The others
either weren't interested, 'too busy', or felt they didn't need to
participate. Well, other than showing the lack of professionalism of our PM
community, it shows the 'in control' attitude of PMs. They are going to 'do
it their way' no matter what, and the history of the organization is NOT
strong on senior management supporting the QA function. The PMs don't like
'visibility' of their project, and it's very hard to find out what is going
on in their projects (other than the monthly status report, there is no
other general forum for QA or Project Office to meet with PMs).
Looking at NCS Level 4 - Unit 5: Apply Skills in Quality Management The
performance criteria are still based on 'motherhood' statements, i.e. 'work
is undertaken according to agreed quality standards'. What standards do you
use if there are no corporate standards except for a very broad PM
methodology or specific technical standards?
Answer: Carl Slate
Here
is a suggestion for you to consider. After reading your second post in
response, I think that you may have some PM cultural issues that really are
not your responsibility, but you may have the capability to make an impact.
Take a look at the Project Management Body of Knowledge (PMBOK). I think
someone else mentioned it, but when I first read it, I was impressed by how
closely it resembled ISO 9000 and quality management concepts in general.
As to
your problem dealing with a quality plan, you are on the right track, your
verification of a project plan and adherence to that plan, with the typical
list of WBS; baselines for cost and schedule; major milestones; key risks
identified; assumptions and constraints; approvals verifying acceptance of
project products etc. But in that process you should also have things like
periodic status meetings with stakeholders or sponsors, periodic review of
scope changes, issues, risk events, along with detailed plans for change
control.
Answer: Bill Duncan
“I'm looking for ideas
about quality assurance for projects and project managers. I am tasked
with developing a 'Quality Plan' template for IT projects and am trying to
determine how to measure the effectiveness of the PM process. Quality
objectives for the 'solution' are fairly easy, as are cost and schedule
objectives.”
Personally, I don't see the definition of these objectives as "fairly easy."
For example, schedule objectives must be more than "complete the project on
time":
-
Does "on time" change
when the product or project scope changes?
-
Do individual activities
or phases have to be completed on time as well?
-
Is there a margin for
error? Is one minute late as bad as one week? As one month?
-
Is working overtime (or
otherwise compromising cost) acceptable?
-
Is reducing
functionality (or otherwise compromising scope) acceptable?
-
Does on time mean
"according to schedule" or "when needed"?
-
Do the stakeholders know
and understand the answers to all the above?
“But what about the
actual process? Most Quality Plans I've come across are based on the ISO
9000 elements, and I am familiar with them. This organization is not ISO
certified and I am not looking for objectives based on test faults, faults
in production, etc.”
You
might want to take another look at the ISO standards. Project management is
a process whose quality can be controlled (in the positive sense of that
word) in much the same way as the manufacturing of ball bearings: develop
your operational definitions, define how to measure the results, and
check-in periodically to make sure both the ball bearings and the
measurements are being done properly.
For
example, you can develop a process control chart for your effort, cost, or
schedule targets. By plotting the actual results, you can show whether or
not the budgeting process is "in control." Whether or not the results are
within tolerances.
“If a PM must have a
Quality Plan for their project, they must tailor it to suit the project.
Other than simply stating the obvious, i.e. the project will be managed
according to the organizational PM methodology, how else do you define
'quality' process for a project?”
Stating that "the project will be managed ..." isn't enough. You have to be
able to demonstrate it.
“Plan and define the
deliverables (Scope, Requirements, Design, etc), number and type of
reviews, then check to see if they followed their plan? Monitor the number
of 'changes' to scope or requirements? Or do you go back to the old
faithful checklist, e.g. is there evidence of WBS; baselines for cost and
schedule; major milestones; key risks identified; assumptions and
constraints; approvals verifying acceptance of project products; etc.”
I was
always taught that quality has two dimensions: conformance to spec and
fitness for use. The checklist approach generally addresses only the former.
The latter requires a little more work as discussed briefly above re
operational definitions:
-
How do you manage the
process to make sure that measurements don't overwhelm the project?
-
How do you know whether
you WBS was any good or not?
-
How do you know whether
key risks were really identified?
Answer: Mary Cahill
All suggestions and advice were appreciated
and I'm overwhelmed at the numbers of you who participated. I have taken on
board most of the advice re references and am purchasing the two standards
recommended. As far as the 'deeper issues' you all were correct in your
assumptions re underlying organizational culture and 'confusion' re roles
and responsibilities. While it is not in my brief to address those issues,
as a QA Manager located within the IT Project Office, I cannot ignore the
current attitudes when working towards overall process improvement.
So far, we have 'reinvented' the project
office for the third time (only this time with employees instead of external
consultants). We are trying to promote the project office as a 'center of
excellence' for PM practices. What I am challenged with is trying to convert
the PM community from their perception of the project office and QA as a
policing or controlling function into an advisory, consultancy, 'we can help
you' function (if only you would let us). I guess I'm most frustrated about
the lack of awareness of process discipline and of having to 'sell' concepts
that I believe should be accepted without such a struggle. I worked in a
large, global IT organization before here, and process discipline was part
of the culture, I didn't have the 'resistance' I have in this organization.
So, to concentrate on the 'process' as many
of you suggested. Yes, I've been more or less following this path. It's just
that even small process 'steps' are taken as 'commands'. I'd love to say we
have a quality policy or commitment to process improvement, but we don't.
After this Quality Plan template for projects, they want me to develop a
Quality Management System
(QMS) for our business unit, even though one
doesn't exist throughout the rest of the organization and it is not a
mandate coming from senior management. I am really resisting this as I don't
really enjoy being set up for failure, even though I do love a challenge.
Without a goal or target to aim for re quality (be it ISO, CMM, or SPICE)
what's the point?
And thank you to Bill Duncan for pointing out
that 'fairly easy' isn't easy at all. I admit I am guilty of
oversimplification.
Back to the main FAQ Repository page
|
