In, “Don’t Let Analytic Bureaucracy Dictate Your Pace” I wrote about how creating more analytic bureaucracy was not the best way to increase analytic throughput. I stated, “You can [increase analytic throughput] by reviewing the bottlenecks in your current analytic organizations for ease of data access, concurrency, support of a variety of modern and traditional languages and tools.”
The feedback I received led me to believe that while there was a general acceptance of the opportunity, much of the challenge is in having a common vernacular in what the analytic organization actually does. Recently I shared this HBR article about how important process is to getting value out of modern and emerging modeling techniques like machine learning and so I thought I’d share the model I use to discuss analytic throughput with my customers.
Visualized above, the BIDMIO model is my go-to tool for raising the conversation above a specific tool or output so that we can discuss what the real bottlenecks are in the process of developing new analytic insights (and getting the most value out of them).