Earlier this month I attended the Gartner PPM and IT Governance Summit in London. Lars Mieritz from Gartner gave a good presentation about key performance indicators and project metrics with the overall message that we should tailor the measures used to report on our projects to ensure they meet the stakeholders’ needs.

That sounds like good common sense, but I think you’d be surprised at how many Project Management Offices (PMOs) use metrics that work for them but that don’t mean anything to the senior management teams or, in fact, anyone who isn’t part of the project management community of that business.

Lars gave examples of different types of project management measures and talked a bit about why these would be relevant for different businesses. Here’s a brief summary.

Demonstrating effectiveness

Some metrics show how effective the project managers are and how good project management is in the business. For example:

  • Number of projects which have been reviewed by stakeholders (an informal customer satisfaction measure)
  • Number of projects with status reports that are out of date (i.e. over 10 days old
  • Number of projects without complete project documentation logged on the system (shows commitment to doing things the right way and with the right decision-support information)
  • Percentage of total project time that it takes to move a project from Initiation through to Delivery (this indicates whether the project has been properly estimated)

Demonstrating maturity

There are another category of metrics that allow a company to track maturity. Here are some of the measures Lars suggested.

  • Number of contractors working on project teams (whether high or low numbers are good depends on your business strategy)
  • Staff churn rate (because staff leaving all the time impacts project delivery as new team members have to be brought onboard and that takes a while)
  • Number of projects originating outside of IT (may illustrate how effectively project management is managed outside of IT and how integrated the IT teams are with the rest of the business)
  • Number of staff who have a formal mentor (shows commitment to professional development)
  • Number of staff who are assigned to multiple projects (again, it might be OK to have a high number, but it could also be a sign of resourcing issues and can flag up pinch points where one resource could become a bottleneck)

Tracking what’s important

If you know what problems your business is facing when it comes to managing projects successfully, then you can target your measures and dashboards accordingly. Lars suggested a few metrics that would help with this – they are very targeted so they might not be appropriate for your company.

  • For companies that have lots of troubled projects: How many projects that had a status of Red have turned Green since the last report?
  • For companies that have resourcing issues: How many projects were fully resourced when they started?
  • For companies where project success rates are low: How many projects complete on time, on budget and with the approved scope changes but no more?
  • For companies where many projects seem to have low benefits: How many projects started with a full business case?

What metrics should you use? Well, that depends on what your stakeholders think is important, and only you know that. What, you don’t know? You should talk to them – they won’t be afraid to let you know what they think and how you could report differently to add more value!

I have come across some stakeholders who don’t know what they want and these people can be difficult to please as everything you present isn’t quite right. They’ll know it when they see it but they aren’t seeing it yet.

If you’re in this situation, think about mind-mapping all your data sources and then thinking about how you could turn these sources of data into measures. It helps if you use software like iMindQ to structure your thoughts and it also makes it easy to share your maps. You never know, someone on your team may see the data in a different way and be able to suggest a couple of new useful measures that you wouldn’t have thought of alone.