Wednesday, June 2, 2010

Creating Proper Metrics

There has been quite a bit of debate around the selection of metrics for project reporting. Most organizations simply look at the "On Time, On Budget" statement and feel that is an acceptable status. Then there are the organizations that try to invest in Earned Value which is a technique that is widely taught in project management circles. The appeal of Earned Value is the reporting of % complete and % of budget expended were not sufficient in telling executives where the project actually stood. Earned Value gives two metrics (Cost Performance Index and Schedule Performance Index) that is engineered to give a quicker and more efficient status to executives. The problem with Earned Value is that it is extremely complex to maintain and most project managers do not have adequate information (like estimated hours to complete and work completed versus hours expended) to create an accurate picture. This is where the stoplight report comes in. Most companies rely on the subjective guess and feelings of the project manager on where the project cost, budget, and quality stand. So what metrics should be reported? The list can vary based on industry and corporate culture, but the following list can help create some metrics to get you started:

• Cost and Schedule: These metrics are still important and understanding where the project stands from a budget and time perspective is still vital information. It just can't be the only information provided.

• Project Plan Quality: How many tasks show a past due start date or a past due finish date? When was the last time the project plan was updated? How many times have the dates slid or moved on the plan? These metrics will show how closely the project manager is managing the plan. Also, by publishing these metrics, it gives the project manager notice that the plan must be managed.

• Project Issues Raised: How many issues have been raised? Too many can be a sign of a project in trouble and too few can be a sign of the project not being managed appropriately.

• Project Issue Close Rate: How many have been closed by the due date? Do issues have a due date? Metrics like this one will make sure that the team is putting appropriate focus on the issues of the project.

• Risks Planned For: How many risks have been raised and planned for? Again, too many shows a risky project and too few shows one that has not been managed appropriately.

• Meeting Attendance: How many meetings have been called and how well attended were they? This helps understand the effectiveness of called meetings. If there are too many or they are not well attended, or too few, this could be a sign.

• Communication Plan Execution: How many communications did the project manager plan during the project and were they effectively communicated? For instance, if they said an issue log would be created and disseminated every Thursday, did they send one every Thursday? This will assist in ensuring that everyone is getting the information that they planned to receive for the project.

• Decisions Requested and Received: How many decisions or issues have been raised to the stakeholders and/or sponsors and how many have had appropriate decisions? This shows the attention that the key people affected by the project are paying. If a decision seems to drag or items requested are overdue, this can put the project at risk.

These are just a few items that can be tracked to paint a clearer picture of how the project is being managed and the level of trust you can put into the reports received. Utilizing a broader metric base will allow you to give a confidence level to the reports and allow you to focus on the proper problem areas. For instance, if a project does not seem to be moving, where does the blame lie? It is easy to simply blame the project manager. However, if the project manager is waiting on the CEO of the company to make decisions and the decisions have not been made, the project manager may be wary to raise a red flag. It is important to make sure that all information is level set so that the most complete picture can be painted for statuses. Then, you truly have the right information in front of you to make the proper adjustment.

1 comment:

Ron Krukowski, PMP said...

Hi Rick,

Earned Value is great when you have all the raw data that you need. And as you mentioned, this data is not always made available to you. Especially when you are an outside consultant.

Capturing the metrics you've laid out here are a great way to keep an eye on things to get a gauge as to the progress of the project. Just a few additional thoughts.

1) Project Plan - along with watching target date slippage, I also look to see %Complete updates. I will trend these statistics from week to week to see if there is a consistent increase in major milestone %Complete. If the %Complete stalls for a few reporting periods or decreases, then I work more closely with that line of business to see what issues they are experiencing.

2) Risks - Ensuring they are planned for. Planned for and not that it is just marked complete on some central risk log but has actual mitigation tasks within the project plan with assigned resources and start/end dates assigned.

3) Change Control - Capturing the number of new requirements, how many of them have been approved by a Change Management Committee, how many days it adds to the plan that was base lined and how far the critical path gets extended are key metrics to track. It also is key to consistently loop this back into the overall project communications so that no one loses sight of the impact caused by the decisions to allow new requirements to enter the project.

4) Resource Tracking Metrics - A key part of the project planning is to have a Roles & Responsibilities Matrix. Tracking resources as they are added in or re-assigned needs to be factored in as well. Time must be allocated for on boarding, task turnover and/or training. Many times the impact to project schedules caused by the changes in resources is overlooked.

These are just a few thoughts.

Thank you for the great information Rick.