How do project controls work?
October 18th, 2002 by GregWe all know the way they are supposed to work. It goes like this — A variance from standard is detected in the reports, the cause for variance determined by managers who take action. Or at least it is supposed to work this way. Can anyone give me an example where this has worked? I mean where the standard was accepted, the variance was recognized as genuine, the root cause was identified, and the action taken made a clear difference.
While we are waiting for the examples to roll in, consider Professor Clarkson Oglesby's observation that people mostly do what management thinks is important. When pressed, he said people knew what mattered by the questions management asked and where they spent money. Two questions come each week straight from management to the crew, "How much did you get done? How many hours did you spend?" In anticipation of next week's "question", people do what they can to make their answers look good. Pipe fitters call the work they do to increase recorded progress near the end of the month, "Show Pipe." The reports look good but real performance drops. The situation is made worse when people select work with the most favorable ratio between effort and credit received. Management turns up the volume as performance declines. They "ask" the questions more forcefully and fire those who can't find a good answer.
Installing show pipe or doing the easy work first ain’t dumb or being difficult. People are doing just what management asked. It is hard to call the results, "unintended consequences" when people respond this way. Let me say it plain — project controls pressurize short-term performance. The fact that the information in monthly reports is rarely used to make management decisions is mostly a good thing. Firing "poor performers" or adding more people rarely turns a troubled project around.
Project controls work in the sense of causing people to pursue local productivity because they regularly repeat the question that appears to most concern management — even when management really wants something else. The effect of these questions cannot be overcome by less frequent questions about something else.
What are we to do? Pay attention to how controls really work. Design cost and schedule reporting systems to cause the action you really want — optimization at the project level- and to provide the information you really need - information to support projections on this job and estimating on the next. Reduce reporting frequency and level of detail to support these objectives. Measure what matters starting with planning system performance. This will tell you how well you are coordinating action within and between crews. Ask about improvements in workflow reliability and the ability to make work ready. Ask the right questions often and in person.
Related Posts
- Project Controls and Work Sampling: A Potent Combination Earlier notes on control were more conceptual. This one recounts how I became so suspicious of project controls. I becam...
- Preview of Linguistic Action Foundations of Lean Construction I've written about linguistic action as a theory that does a better job at explaining why projects succeed and why they ...
- The Future of Project Controls I'm somewhat hesitant to write about this. This morning I received an email newsletter that included advice on How to g...
- Safety Book is a Winner Safety Thursday Work Injury Free I've continued my search for a few good books on safety. Thank you to those readers ...
- Capacity for Language Introduces Uncertainty There's a book or two (or three) on this topic. I'll make a few points. Perhaps you can add to them with your comments...
