
To manage an analytics project, one must first have a map of the terrain. For decades, organizations have relied on rigid frameworks to guide the journey from "data" to "insight." However, as the tools and roles in our industry have exploded, these older maps have begun to show their age.
Before we introduce a modern framework, it is essential to understand the two dominant models that have shaped the industry for the last twenty years: CRISP-DM and the Business Analytics Model.
The Cross-Industry Standard Process for Data Mining (CRISP-DM) is the grandfather of analytics frameworks. Conceived in 1996 and released in 1999, it was born in an era where "Data Mining" was a specialized, expensive activity often performed by a few PhDs in a basement using heavy workstations.
The Context of its Birth:
CRISP-DM evolved in a business environment dominated by proprietary, expensive tools like IBM SPSS Modeller (formerly Clementine) and SAS. The process was designed for a "Waterfall" world: you gathered requirements, you mined the data, you built the model, and you handed it over.
While CRISP-DM focused on the technical execution, the Business Analytics Model (as detailed by Laursen & Thorlund) focused on the value chain.
This model emphasizes that analytics is not a straight line, but a continuous loop between Strategy and Execution: