Friday, January 6, 2017

Project predictions

Most of the formal measures we have of project performance (including progress) are backward looking. They report what has been done, sometimes when it is too late to bring corrective action to bear if performance is below expectation.

There is plenty to do to avoid project mistakes: credible baselines, to borrow from Glen Alleman's blog, are a big part of it, as are achievable requirements, risk-adjusted planning, the right team, and so on. That's essential to project performance.

All well and good, but how do we look ahead?

I recently received a newsletter from Charles Pellerin and this turned my mind to the project social climate. Like any business undertaking, projects are delivered socially: that is by people working together to achieve a common goal. Social climate, the sort of thing that Charles writes and practices on, is a large part of determining project success. There are social climate instruments that in expert hands could provide a useful insight into project outcomes. But a wise PM can also maintain observation of social climate: what issues, risks and decisions are coming to him or her, and how quickly are they attended to: a good rule of thumb.

In $300m major urban renewal project that I directed some years ago, I used 'aged decisions' as my rough guide to the performance of the team (reaching across a number of government and private sector players). It also gave me something tangible to open conversations to prompt performance when a decision remained unmade for an excessive period.

Another way of looking at it, is to consider promise reliability: measured by promises delivered. Following Hal Macomber's work in the old Reforming Project Management blog (some of the material might be found on the Lean Construction website); the tracking of promise keeping is an indicator, a robust one, in my view, of project delivery reliability.

There are others as well. In a contracting environment contract commitment against plan is a useful forwarding looking metric, for example.

No comments:

Post a Comment