One of the favorite jobs I have had was running the CapStat performance management program for Mayor Fenty of the District of Columbia. CapStat was the performance management program Mayor Fenty launched to assess the quality of services delivered by the $10 billion local District of Columbia government. CapStat was inspired by other “Stat” systems that had proven successful in local governments, such as CompStat in New York City and CityStat in Baltimore. At the core, CapStat was responsible for finding opportunities for improvement in District government operations and then hold people accountable for making those improvements through rigorous action tracking.

Here are the key elements of CapStat that enabled that.

The CapStat Team – The CapStat team was a small group (usually four people) of analysts from a mix of backgrounds in management consulting and information technology. There core skills were in problem-solving, analysis and presentations.

The CapStat Meeting – About once per week, the CapStat team would organize a one-hour meeting with the mayor. Each meeting would be on a single issue facing the city and would include the heads of all the agencies involved in that issue. For example, a session on graffiti might include the heads of the Department of Public Works, the Metropolitan Police Department, and the Department of Consumer and Regulatory Affairs. One of the agencies would be determined to be the lead agency. That agency would open the meeting and then it would become much more of a question and answer session led by the mayor.

The CapStat Presentation – The CapStat team would work with the involved agencies to identify the key information the mayor needed to see on the issue. Together they would gather the data, analyze it and then summarize it in PowerPoint slides. The deck would typically total about 15-20 slides, with 20 as the max to keep focus. If the team did its work well, there would be a slide relevant to every question the mayor asked.

The Action Items Memo – During the meeting, I would capture any actions the mayor asked to be done during the discussion. At the close of the meeting, I would project those items on to the screen for all to see and add a “who” and “when” to each. The members in that room could then ask to edit each until we had a final version. Then I would include those into an Action Items Memo and send them to all involved after the meeting.

The CapStat Calendar – Many issues requiring a CapStat session were recurring items, like readiness for snow removal operations or reopening of schools and parks. We would schedule those meetings with the appropriate lead times to give the mayor the opportunity to find out about preparedness for those types of events.

The Action Tracker – We entered all those Action Items into a Quickbase database we called the Action Tracker. Quickbase is a web-based database where we could share views to the people in the agencies. Quickbase turned out to be an invaluable tool to track completion and manage deadlines.

TrackDC – We integrated the Action Items from Quickbase into the TrackDC dashboard, which had a page for each agency showing the status of the Action Items as well as key operational data (e.g., spending versus budget). We set it up so the agency heads could see their own pages too to help them track their agency.

The “Stock Price” Score – We created an algorithm to calculate a 0-100 score of agency performance based on the data in the dashboard. We had higher weights on the things we most valued such as completing their annual initiatives, hitting their Key Performance Indicators, completing Action Items requested by Mayor in CapStat sessions, spending on pace with budget, and their responsiveness to correspondence from citizens and our “mystery shoppers”. The score would go up or down daily based on new information and aging against deadlines. The agency could click on their stock price to see exactly what factors were driving it below 100.

Track.dc.gov – We created a public version of these dashboards and made them available to the public at Track.dc.gov.

As a whole, the CapStat structure and tools created a powerful management infrastructure to help ensure the required improvements in District local operations happened. I presented an overview of it to senior city staffers from about 20 other large cities at a Harvard University Kennedy School of Government conference in 2010 and several cities asked for help in applying parts of the system themselves. Contact me if you would like to see a copy of that presentation.

You may also like our other articles, such as: A Case Study in Management Consulting or 12 Best Practices from Elite Executive Assistants