How to fix Stat
The “Stat” performance management model features interactive conversations focused on the evaluation of trend data. In theory, these discussions produce insights that can improve operations.
However, the model has begun to languish and face dilution as the initial hype has subsided and analytical software has become ubiquitous. Stat’s challenges in its hometown are indicative of the need to calibrate the model to avoid staleness.
Stat can still be a tool for meaningful performance management by focusing on its strengths. Specifically, Stat excels at interactive deliberation about resource allocation decisions, especially when there are still conditions of uncertainty even after evaluating available data.
On the other hand, Stat is no longer a suitable routine for discovering individuals or teams that are “positive outliers” that can serve as the foundation of improving “the way we do things.” The widespread availability of software tools for data analysis allows for that discovery to be automated. Time and other resources should shift towards properly documenting the technique of the positive outliers, as well as teaching it to other team members.
Similarly, new software tools allow the automation of operations optimization. These tools greatly simplify the ability of an enterprise to make automated recommendations (such as scheduling, routes, project sequence order, etc) the time-effective starting point for a division’s performance management. Stat discussions are a relatively inefficient way to make these choices.
How do you know if a Stat-style process is still right for you? Here’s a quick checklist :
- There are clearly identified resources (e.g. patrol cars) that need to be allocated at a quarterly or higher-tempo basis.
- There are observable outcomes to optimize for that are impacted by the resources (e.g. reported crime).
- Full automation of resource allocation optimization is not viable due to data quality issues, substantial uncertainty about future conditions, or hard-to-define constraints such as “equity” or “stakeholder perceptions”.
Performance management is too complicated for a one-size-fits-all approach, so adjust these recommendations to your organizational context. Based on our experiences with this format, Stat can still serve data-driven leader when it is deployed to leverage its strengths.