Project triage

Medical image How do you quickly figure out where to put your time and effort when troubleshooting projects? Mark White, a partner at WM Partners, explains how timeline charts help you cut to the chase.

Triage ER-style

In the late 90s, doctors in the Emergency Department at Cook County Hospital in Chicago (the inspiration for the hit TV series ER) reviewed the way in which they assessed suspected heart-attack patients. The reason for this was too many patients were being unnecessarily triaged into intensive care, taking up valuable resources that should have been used for sicker people. In addition, doctors were spending a lot of time assessing each patient which reduced their overall throughput.

Doctors were instructed to gather less information on their patients and zero in on just four critical pieces of information about patients suffering from chest pain; ECG results, blood pressure, lung noise and existing unstable heart disease, while ignoring everything else associated with the patient's wider medical history. This approach resulted in a considerable improvement in efficiency without compromising patient safety.

Greater detail doesn't always bring greater clarity

Over the years, variously as a programme manager and project business sponsor I've sat in many progress meetings thinking:

'Hasn't that milestone slipped previously? How can I be sure that the project isn't in permanent drift?'

or,

'Is the project team being over- optimistic about what will get delivered next period? It feels like delivery is forecast to accelerate, but I can't see what has changed to make this the case.'

Generally, I've been provided with lots of detailed information about the present and immediate future (the ubiquitous 'this period/next period' reporting model). In addition there is typically a mountain of supporting material including risk registers, Gantt charts and, if you have a creative project manager, heat maps.

Often in these situations there is a temptation to ask for more detailed information from the project team to justify their estimates and delivery dates. Counter-intuitive as this may seem, this is not the right thing to do. Instead, you should use project triage techniques to zone in on the hot spots before delving into the detail.

Triage project-style

So, what's effective heart patient triage got to do with quickly spotting problems with projects? They both require an initial focus on a small number of metrics before deciding on what detailed actions can best address the situation.

One important metric to follow is milestone slippage. Status reports typically only cover a comparison between the last and next reporting period - any trending information on delivery dates is lost. However, if you look at how key milestone dates change over time, you can use this to identify delivery trends that are otherwise obscured.

The timeline chart

The timeline chart is a simple technique that helps spot milestone trends across the lifetime of the project (or phase of a project if you wish). It shows how the planned delivery dates for key milestones have moved as the project has progressed. Essentially, you plot each milestones predicted date over time. Here's an example.

Planned delivery date - the delivery date for the milestone as estimated on the Reporting Date

Reporting date - the date on which the delivery date was estimated.

What does this tell us?

Each coloured line on the chart represents a milestone.

Sign-off spec  was initially forecast in January to deliver in March. It delivered on time in March. There has been no slippage - the dates originally estimated for delivery have been met.

UAT starts was initially forecast from January to March to begin in July. UAT actually began one month early in June.

UAT ends was initially forecast from January to July to finish in September. From August onwards, the milestone has consistently slipped a month each month. This milestone is in serious trouble and is where extra attention should be focused.

A note about milestones

This approach is only as good as the milestones you measure. The trick is to choose only the truly significant ones. Don't choose too many or the detail starts to crowd out the trend information.

Each milestone needs to be relevant to the business, explicitly defined and independently measurable. So 'user acceptance testing completed' is better than 'code complete'.

For a small number of projects, office productivity tools such as MS Excel or PowerPoint, or paper are the way to go. We've provided a simple template in MS Excel which can be downloaded from the WM Partners website for this purpose. To track milestones in this way across a large portfolio of projects, and have the results available to you for comparison across the portfolio, you may wish to consider automating this through tools.

WM Partners provides Project Recovery, IT Strategy and Project Delivery services. For more information contact Mark White.

© 2008 WM Partners LLP.

April 2008