9

I am the Infrastructure PM on a large project, working as part of a team of PMs with different responsibilities. These include Development, Testing, Environment Support, and Implementation. My remit is limited to delivering the hardware and middleware components required to support the application.

The main technology that I have to implement in the current phase of the project is IBM WebSphere. There is no new hardware: as far as infrastructure is concerned, we are simply adding new WebSphere components to the technology stack. I have access to two specialists in the WebSphere team, as well as other ancillary specialists for the other (less prominent) technical changes that I have to make. The problem is this: each of the other PMs also needs the WebSphere guys to support their parts of the project.

We work together to align our plans, and everything looks great on paper, but the fact is that we end up scheduling 100% rather than 70% or 80% of their time, so no slack exists. And then reality kicks in. Something goes wrong: one of the tests takes longer than expected, or one of the test environments falls over, or we need to deploy new code more frequently than expected, to resolve defects in test. Any of these can mean that our plans for these scarce resources immediately fall over.

My question is this: What can I do to encourage my fellow PMs to build in a lot more slack into the project, and stop filling every minute of every day with planned work? No matter how much I try to do so, someone comes along with a new task, sees a couple of hours of apparently unoccupied time, and lays claim to it - with predictable (and disastrous) results.

Improve this question
0

4 Answers 4

Reset to default
5

In the short term can you change the business rules for the time of the WebSphere SMEs so that you can't allocate them 100%? For example, if the current business rule is that they work 8 hours per day from 0900 to 1700, change the rule so that they work 6 hours per day from 1000 to 1600. Or block off "me time" from 1100-1300.

In the longer term can you get the other PMs properly trained on project management and team leadership? It sounds like they have some problems with basic concepts like:

  1. Resources aren't 100% efficient... 60-70% efficiency is more what you should plan on
  2. Switching between tasks/projects reduces efficiency... coordinate with the other PMs so that a resource only has to work on one project per day
  3. If you treat someone like a piece of meat they will act like one i.e. they will not contribute to their full potential... or they will turn around and walk away and really leave you in the lurch

That being said, in the long term the SMEs should also be educated to push back on you and the other PMs when they are overtasked. They have to be part of the solution as well as the PMs.

Improve this answer
1
  • Some excellent practical suggestions here, that I think we could work towards. I like the idea of one project per day per person, which would certainly help, and you are right in your implied criticism... I am probably as guilty as the rest of the PMs!
    – Iain9688
    Commented Jun 22, 2012 at 19:20
6

This looks like a utilization vs throughput situation. Managers who believe in utilization will always schedule 100% of the working hours of their employees, and that's why they don't allow any slack in their systems. (on the other hand they live like this, unfortunately, and bring this standard to the workplace). I would try to explain the difference between utilization and throughput.

My advice is to find bad examples from the workplace, when overburden employees made mistakes. I read about toyota that they didn't run their machines 100% all time because they were afraid that the machine would get broken. This is one of the waste they defined.

When I'm talking about this topic, I'm using the following arguments:

  • 100% coverage means that the colleague won't have time to be creative
  • 100% coverage doesn't mean that the project will be finished. It means that the colleagues will work the whole day. On the other hand, keeping the throughput in mind will actually make a difference
  • slack time != buffer
  • having slack time doesn't mean that the colleagues won't work as much as they should be. Based on my experience they use the time wisely. Those who don't they find the way not to work even when the 100% coverage is in place
  • 100% coverage makes us inflexible and slow. There is no place for change or fast reaction
Improve this answer
1
  • Excellent observations, and the lack of creativity (or even time for proper discussion of debate about the best approach) is squeezed out by the regime that we have adopted. I know that we could find good examples even within our own project where a bit of time spent on thinking would save much more time when it comes to doing. In fact, we actually did a bit of proper thinking today, and it has saved us a lot of effort downstream, so thanks for pointing that out.
    – Iain9688
    Commented Jun 22, 2012 at 19:29
2

While I agree in part with most of the other answers, the two things missing for me are 1) the (apparent) lack of a Program Manager, and the lack of anyone responsible for resource allocation/planning.

On a project with multiple PM's there's no Program Manager or PMO to avoid this situation?

If not, then I would suggest that you and the other PM's meet and decide on one of you to, not act as the PgM, but to at least be responsible for resource allocation. All of you plan your various parts of the project, and then pool the plans to find the overlap and over-allocation of resources so that you can avoid this.

I know you said you work to align the plans. I'm talking about aligning them, and then specifically looking for either over-allocation, or just allocated at 100%. Given your situation, this is as much risk management (for all of you) as it is resource allocation.

Improve this answer
1
  • Good comments Trevor. We do have a programme manager and a PMO, but their focus is more on combining the plans that we provide (without doing much to identify resource conflicts) and then monitoring whether we are achieving our stated plans, than on helping to develop viable plans in the first place. You have certainly pinpointed a weakness for our programme. In terms of risk management, we do have the resource conflict listed as a high impact, high probability risk, but we are expected to manage this particular risk between the members of the PM group. Good insights!
    – Iain9688
    Commented Jun 25, 2012 at 19:54
1

How are you building your estimates? Deterministically or probabilistically? Do you take advantage of simulation software to estimate? There is no task in the world that take x units of time to finish. Instead it takes between x - something to x + something to finish. And the variables that cause you to finish somewhere early or somewhere late are many, random, and largely uncontrollable.

If you are not conducting this type of analysis on your schedule, or your budget for that matter, then you are not really aware of the risk you are taking on and, therefore, not prepared to cope with variables that are unfavorable to you by way of contingency.

Improve this answer
3
  • It's a fair question, David, and the answer is that we don't estimate too well at all. It tends to be a case of trying to fit the work into the time that someone has allocated to us, rather than building a project estimate from the ground up. So... neither probabilistic nor deterministic, but more of a "land grab" whenever we can find a bit of time in someone's schedule.
    – Iain9688
    Commented Jun 22, 2012 at 19:24
  • If you are not estimating and you don't know the probabilistic density of your project, adding slack or using some rule that a resource is something less than 100% will not help you. Maybe for one or two, but eventually people will land grab that up anyways. You need to beef up your estimating capabilities.
    – David Espina
    Commented Jun 22, 2012 at 20:02
  • 1
    I can't argue against that. Poor estimating is a problem across the whole project, and although we hide it behind all sorts of fancy excuses, that is probably at the heart of a lot of our difficulties. I think I need to sharpen my estimating skills, investigate the tools that can help me, ... and also try to influence the rest of the team too!
    – Iain9688
    Commented Jun 22, 2012 at 20:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.