The lord of the metrics

So I’ve been thinking about my earlier post on “Metrics: One Rule to Rule Them All.” It argued that each government agency should have just one metric to guide its performance, very much like the President has one approval rating that we use to gauge his or her success. The idea of having one or just a few overarching metrics is a common practice in leading private companies like Apple and, of course, for politicians like the President, but it is the opposite of what is done in the agencies of the U.S. Federal Government.

But what would that look like? It would vary by agency, but in the U.S. it would look something like this:

  • U.S. Department of Veterans Affairs: The Veteran Satisfaction Index, polling the U.S. Veteran population (appx. 23 million people).
  • Social Security Administration: Recipients Satisfaction Index, polling the benefits-eligible population.
  • Department of Homeland Security: Safety Perception Satisfaction Index, polling the entire American population.

Just like the Presidential polls, these should not be limited to just those who actually come in contact with the agency, but everyone who may come in contact with them. Because those people who do not come in contact but are eligible to matter as well. They could complain to their Congressman that the agency is unfairly shutting them out — and it may be true. So they need to be in the poll.

These indices would need to be national and rolling. They should be modeled after electoral polls, which are well established and effective. They would also need to be detailed enough that you could dissect the data based on the geographic location of the pollee and the particular service they interacted with, similar to how electoral polling data can be dissected.

Some cool benefits of this new approach include:

  1. Solve the metrics in-fighting problem that agencies have, and unifying them together toward a common goal.
  2. Providing leading indicators for future service failures (and preventing bad press as a result). Having leading indicators that can predict trouble in your organization’s future service delivery is a critical best practice used widely in the private sector that Federal agencies systematically lack.
  3. To expand on #1, it allows agencies to better connect its staff functions’ performance (HR, IT, acquisitions) to actual service delivery, which are generally measured separately and disconnected from each other.

Original post

Leave a Comment


Leave a Reply

Carol Davison

Metrics change with strategic plans which change every five years or so.

Also your metrics are based on customer values and perceptions, not the agencies mission. For example Social Security only wants to provide deserving people what they deserve and in a timely manner. I might think that they should give me scads of money because I’m old. I might be wrong. I might complain very vocally and very effectively.

I’m not particularly interested in our agencies’ popularity, I’m interested in them achieving results.

Also we need several metrics. At least one for line and another for staff elements. At Commerce we . faciliate international trade, foster science and technology and protect intellectual property, protect the environment and achieve organizaitonal and managmeent effectiveness.