One Number, One Decision
The most useful performance reports tend to have very few numbers on them. When I managed a population health program, one key measure was the share of members receiving in-home visits. A measure like that moves only when the people looking at it can act on it the same week.
It is common to see a dashboard with dozens of measures and a meeting that ends without a decision about any of them. The software rarely explains the difference. More often it comes down to whether each number on the screen has been given a purpose.
Begin with the decision
Dashboards are often assembled from whatever data is easiest to pull. The better ones are designed in the other direction, starting with a decision someone has to make. A care coordination lead needs to see where follow-up is breaking down. An operations director needs to know whether a new workflow is shortening delays. A community program needs to know whether the people it is reaching are the people it was funded to serve. Those are three separate views, and a single dashboard meant to serve all of them tends to serve none of them well.
Before a measure is added, it helps to ask who will act on it, how often they can realistically influence it, and what they would do differently if it changed. When nobody can answer, the measure may still belong in a periodic report, but I would keep it off the page that guides weekly decisions.
Show the work next to the outcome
A visit rate is an outcome measure, and by itself it can only show that a program is behind. It becomes useful beside the work that produces it and what that work costs, such as visit attempts made on schedule and miles driven. Pairing the two shows whether a higher visit rate is coming from better planning or simply from asking staff to spend more hours on the road.
The same reading applies to most dashboards. When an outcome is weak and the process measures are also weak, the issue is usually execution. When the process is being followed faithfully and the outcome has not improved, the underlying approach deserves a second look. When the outcome improves while staff burden rises, the gain is unlikely to hold. A single figure cannot distinguish among those situations, and each one calls for a different response.
Look underneath the average
A steady overall score can conceal one site, one language group, or one referral source where things are going poorly. AHRQ’s national quality and disparities reporting is organized around that idea, and it applies equally well inside a single organization. It helps to break results down only as far as a decision requires, since an unlimited set of filters can become a way of postponing the decision.
Record what was decided
One small habit does a great deal to keep a dashboard useful. Beside each indicator that is off track, write down what was decided, who is responsible, and when the group will review it again. The next meeting can then begin from that note, and the team does not spend its first twenty minutes rediscovering the same problem. It also helps to move a measure off the main page once it has been stable for several months, because a dashboard that only grows eventually stops being read.
A measure earns its place when it starts a regular conversation about who has not yet been reached and what the team plans to do next. That is a reasonable standard for every number on the page.
Sources
About the author
Sherry Sayari, MPH, is an Operational Standards Evaluator who has led population health, community outreach, and social services programs. More on the About page.
