The last several posts have been around PM 2.0 and what that means, if anything to the project management space. It's time to remind all those claiming to have the next big thing in project management of the core elements of a success project.
First let's look at the sources of project failure. There are many of course. My particular point is view is from the Program Planning and Controls paradigm. There are numerous, probably too many to count reasons on the "softer" side of projects. And likely as many Tools reasons. .While people are certainty the most important aspect of a project, the tools are a distant 3rd for reasons mentioned in the past.
Let's look at the process needed to keep a project on track. Or another view, what things can derail a project.
Let me Count the Ways
- Inattention to budgetary responsibilities – if you're not watching the money, you'll run out and not know it. I've heard in the past that budget is not watched on many projects. I suspect this may be the case in immature organizations. But ask this. "do you watch your budget?" If you live pay check to pay check, you're implicitly watching your budget. Then why on God's Green Earth would you not watch the budget on a project. Especially if it was someone else's money?
- Work authorizations that are not always followed – as a project team do we work on anything we want to work on at any time we want to? Hopefully not. Working on the right things in the right order is part of project management.
- Issues with Budget and data reconciliation – when we're working on something, do we know how much we're spending? How much did we plan to spend? Do they match in any way?
- Lack of an integrated management system – an integrated management system could be 3x5 cards on the wall or a full blown SAP program management enterprise suite. Doesn't really matter in principle. But have something. Don't show up evey morning wondering "what should I work on today?" "What did I get done yesterday?"
- Baseline fluctuations and frequent replanning – are we bouncing all over the place when it comes to cost and schedule? "today it'll take 3 weeks to finish." "Nope, today it'll take 6 weeks." "Next week we discovered it'll take 12 weeks." Come on guys how ling is it going to take and how much is it going to cost. These number by the way estimates. They need variances and confidence intervals in order to be credible.
- Lack of predictive variance analysis – if I can't forecast future performance in some way, it'll be hard to tell when we'll be done. Agile uses velocity. Big programs use Earned Value. Use something. Don't just sit there and look stupid when asked "any guess about when we'll be done and how much it'll cost when get there?"
- Progress not monitored in a regular and consistent manner – "how we do'in?" "Oh I don't really know, we're taking this one day at a time." That's a possible answer for some situations, a death in the family, a recovery from illness, or a 100 mile loop on a road bike around the Colorado mountains. But probably not a good response for a project manager. Know how you're doing. Know that of you keep going at the rate you're going you'll arrive at some projected time. Just like looking at the speedometer on the road bike. "If I keep peddling at this rate, I'll arrive in Leadville in two hours."
- Lack of internal surveillance and controls – who's watching from the outside? Who agrees things are going according to plan. The notion that there is a plan is the first step. But let's assume there is a plan, who gets to say we're moving along according to plan?
- Managerial actions not demonstrated using performance measures – we discover we're late or over budget. Is there a management decision to do something about it? It doesn't have to be formal "management," but is there a decision by anyone to do something different. Or are we just containing blissfully down the road to the train wreck?
So now let's assume, and it might be a big assumption that we want to do something about this pending Train Wreck. What would we do?
Staying Out of the Ditch
Here's a set of answers taken for a large set of answers. The larger set is the 32 criteria of ANSI-748B Earned Value. Now Earned Value is not being recommended here. But most of what's in the EV standard is about "good project management." Not actually EV.
- Decompose the project into manageable units of work held in the Work Breakdown Structure (WBS) that states what DONE looks like.
- Assign responsibilities for the accomplishment of the each element of work at the appropriate management level using the Organizational Breakdown Structure (OBS). Join these two structures into a Responsibility Assignment Matrix (RAM).
- Create a schedule which identifies the work activities, milestones, and their dependencies for each intersection in the RAM connecting accountability with deliverables. Adhere to the schedule – last starts usually mean late finishes.
- Assign resources to the activities, determining the cost of these resources, producing the time phased budget for the project. The result is knowledge of who is accountable for the deliverables and what is their budget?
- Identify the objective measures (physical percent complete is preferred) for all the identified work – what does DONE look like and how are you going to measure progress toward DONE in units meaningful to the project?
- Establish a change control procedure for the Performance Measurement Baseline (PMB) – assuring changes don't unfavorably impact the plan.
- Formally release the work packages for execution by the assigned resources during the planned period performance – do the work in the right order.
- Record and accumulate the projects progress through cost and schedule performance measure – assure you're measuring physical percent complete.
- Use the cost and schedule variances to develop an Estimate At Completion (EAC) and Estimates to Complete (ET) for all future planned work – use data to "manage."
- Take managerial action to compensate for past deviations or correct the planned work to meet the approved Budget At Complete (BAC) on schedule – "manage the project"
- Track and manage all changes to PMB. Past performance is a good indicator of future performance – knowing how you got to where you are now, helps knowing where you'll be going to be the future if you don't make corrections to your course.
