Understand the difference between describing what happened, explaining patterns, forecasting what might happen next, and using a model to recommend action.
Analytics exist on a spectrum, and most departments can create significant value by getting better at the first two steps before going near a model. Descriptive analysis asks what happened — incident volume, response times, workload, overlapping calls. Diagnostic asks why it might be happening: travel distance, hospital turnaround, unit availability, time-of-day demand. Predictive asks what is likely to happen, forecasting demand or estimating future resource needs. Prescriptive asks what action might improve the result. Prediction is not certainty: a model finds patterns in historical data under assumptions, and should never be treated as a guarantee. Be especially careful where a model influences deployment, personnel decisions, patient care, inspection or enforcement activity, or anything safety-critical.
If you read one thing on this subject, read this.
The national system built to support empirical decision-making and, in time, predictive analysis.
Why it matters
Sets out the intent directly, including that the analytic capability is still being built.
How agencies get their own data back out for comparison and analysis.
Why it matters
Descriptive analysis of your own agency is the step most services skip on the way to a model.
Related guide on the Hub.
Everything above was published by someone else, and is here because it is the clearest treatment of the subject we could find and verify. The Hub’s own guide to this topic is still being written. If you know a better source than the ones listed, that is worth telling us before it is.
Send a note