How It Works

Through historical analysis, we see where potential innovations in your company may be hiding on a given objective. Through working with our clients, we begin to discover what ideas could drive better performance. Using Statistical Design, we then test all of these ideas (aiming for 20-50+ individual “interventions,” or changes) examining how they work together to attain objectives. In doing so, we provide an implementation model, based on testing and data in the live business that leads to competitive advantage. Tests are quick and effectively provide a wealth of information and insight into our client’s processes.


However, our methods are stronger and more insightful than predictive models alone. Predictive models derived only from historical data prove comparatively weak in the face of achieving true innovation. While sometimes a predictive model will solve a problem, Statistical Design, provides a proven method to solve complex problems.



We do data mining and predictive models initially, and these efforts show us what to explore more thoroughly in the full Statistical Design. Through running all potential interventions aimed at achieving innovation through a Statistical Design, the end result is a simple implementation model that tell us, with remarkably high levels of accuracy, what performance can be expected based on real-world results and actual implementation.

Predictive models do not factor how the real world will react to proposed changes. Our models accept real-world conditions as the testing environment. In this regard, the world is not perfect- it isn’t constant, and it won’t hold still while we attempt to figure out the best route to competitive advantage. With the ability to find and quantify interactions where a pair of interventions work synergistically--or acrimoniously, Statistical Design lends us even more insight into improving performance. There isn’t another solution available that lets us figure out complex problems as we encounter turbulence.


“The long range contribution of statistics depends… on creating a statistically minded generation.”



--Walter A. Shewhart