We are having difficulties deciding how to measure project status. We cannot use earned value because we do not use project time reporting - we use the old % complete by task. Our executive team wants someway to know whether a project is on time or starting to show signs of problems. We tried using percent of total project complete, number of milestones and how many were complete, and percent complete of each project phase. Anyone have any suggestions on what has worked best for them? Saving Changes...
Sort By:
Michael WoodProject Manager / Business Analyst / Business Process Improvement Guru| Independent ContractorGig Harbor, Wa, United States
Jill, there is nothing wrong with the percent complete method. The real issue is projecting the future of the project and how to demonstrate that the status reports are accurate. You might want to consider the following approach given you really only have proof of what has been completed. Break tasks into single person efforts. Each week, have people indicate the hours they have left on a task and the percent they beleive the task is complete. Also have them commit to completion dates (promises if you will). Then, via your Project tracking tool, put together a resource utilization of each person's total commitments of time for the next month. If it comes out to more than 120 hours (6 hours of true productivity per day), you are probably going to have schedule slippage. Publish to management, a weekly report that shows tasks that were due to complete the prior week, what was actually completed, corrective actions to resolve past due tasks, tasks to be completed next week and obstacles you can foresee that could impact on time - on budget completion of the project. This will show a linkage from week to week and that the process is being managed with diligence. Finally, hold your people to thier promises. Publish them on the master project task list report. Don't dwell on tasks that are done, only dwell on the future tasks to be completed. Remember, you cannot change the past, you can only influence the future. Hope this helps a little. Saving Changes...
I almost gagged when I saw Michael start out with saying that there isn't anything wrong with percent complete. Then I read on and realized that he suggested also coming up with an "estimate to complete" as well. Percent complete tells you what one thinks was done, but doesn't say anything about the status of the project as it says nothing about what remains to be done.
In the Critical Chain world, we rely on reported "estimates to completion" of tasks to develop a cumulative view of project buffer consumption. The Project Buffer is developed to protect the project's promise -- the due date -- and if it is consumed too much, or being consumed too quickly in comparison to completion of the critical chain, then we have a potential issue to address.
For simple status reporting in the critical chain world, we simply talk about percent of critical chain complete to tell us where we are, and the condition of the buffer, usually in a green, yellow, or red coding. Green means "life is good", yellow suggests that a contingency plan be developed for implementation if the buffer deteriorates to the red, and red should only be reported if the contingency plan fails to work. (There are also alternative views of buffer management that compare buffer consumption to chain completion, or buffer remaining to a recalculated buffer needed.)
But it's all based on frequent and regular reporting of "estimates to complete" of active tasks.
Saving Changes...
Michael WoodProject Manager / Business Analyst / Business Process Improvement Guru| Independent ContractorGig Harbor, Wa, United States
Scared you a bit Frank? I do agree, that what is left ahead is far more important than what has been completed. In my more complex projects I used the percent complete and the effort remaining to forecast estimating errors and to recast the plan. For example, if a task was estimated at 20 hours and after 10 hours is 20% complete, then there is a reason to conclude that the task will take another 40 hours to complete. This compared with the estimate to complete can be very revealing. A reality check of sorts. But this technique does require capturing of effort to date. Saving Changes...
Thanks for the help, Michael. I have since been talking to other PMP's and referring to some good books and your suggestions are right on. I need to compare what was completed with what should have been completed and estimate where I will be next review period. Looking at individual tasks is necessary for a weekly review but I'm thinking I could just look at milestones for the monthly management review. The key is making sure the plan has frequent and meaningful milestones (tied to a specific deliverable - that way there can be no question about whether it is complete or not). Let me know if you have any other thoughts. Again, many thanks. Saving Changes...
This dialogue got me wondering, other than purely physical projects, like construction (where one can go out and count the number of blocks laid) how valid can any percent complete assessment be made. I've supported clients primarily in the product development and software arenas and other than the very few tasks where there are valid estimates of lines of code associated with a task, the idea of percent complete feels bogus to me.
Again, I'd rather have the resouce look forward and simply reestimate the time to complete for its daily/weekly reporting.
I remember the old saw about software development: "90% of the project takes the first 90% of the time, and the final 10% of the project takes the other 90% of the time!" I read a very well-reasoned article a couple years ago that said, if you broke down the work far enough (no more than 2wks/task), you could get a reasonable approximate of earned value by simply asking "is the task started?" and "Is the task finished?" If it was started, then assign it 50% complete. If it was finished, assign it 100% complete. This obviously requires a good wbs, but in an environment where you don't have accurate time accounting (and maybe DO have individuals who will fudge on their progress when it comes to reporting time), it's a good progress indicator. I'm pretty sure it was in an issue of Project Management Journal, maybe 1998 or early 1999 Saving Changes...
Michael WoodProject Manager / Business Analyst / Business Process Improvement Guru| Independent ContractorGig Harbor, Wa, United States
Thanks Jill, I did have a few more thought. First, you might want to tie major captial outlays to the plan. I find that management likes to see that sort of thing. Also, instead of approval sign-offs, which people are reluctant to give - that accountability thing I guess, try calling them Authorizations to Proceed. Few people will want to hold up a project - that political thing you know. Also, when you give management the latest detailed project plan update I suggest only showing them the open tasks. I shade the entire task lines with the following colors to make it real easy for them to scan the plan. Green = on schedule. Yellow = behind schedule but will be back on track next week. Red = behind schedule - need management intervention and support to get resolved. And my favorite Blue = Ahead of schedule or below budget. You can pick your own colors of course. And if you really want to impress the top brass, be able to articulate the lessons learned that can be applied toward future projects. Remember there are lessons learned from problems and successes. It is very impressive to learn from success - demonstrates it can be repeated and not just dumb luck. These are little things I know, but trust me, sometimes its the little things that can make the difference to your credibility. Saving Changes...
Michael WoodProject Manager / Business Analyst / Business Process Improvement Guru| Independent ContractorGig Harbor, Wa, United States
Frank, as usual I agree with your assessment and opinion. By itself, percentage complete is useless if it cannot be verified by direct observation. The only value I have found in it is for assessing the team's ability to forecast the effort needed to complete tasks. By tracking and feeding back to them their forecasts to actuals, they can hopefully become better at the task. It also helps develop an estimating error factor for the future. But looking forward is always the way to go. You can't resource a percent complete. But you can resource what is left to get done. Afterall, one of the main imperatives is to align the project plan with reality. Saving Changes...
Jill Henery Tracking project status without earned value can be tricky, but there are other effective ways. You can use milestone tracking, planned vs. actual completion dates, or a traffic light (RAG) system for risk visibility. Another option is tracking key deliverables and team progress through weekly check-ins. Have you tried setting clear phase gates with defined criteria for completion? That can help flag issues early.
Community Champion
