The story of a modelling decision…

User research consistently pointed to an important context: investors and fund managers think about their domain through the axes of strategy, sector, and region.

Existing data, editorial content and event objects all stored strategy, sector and region as facets of objects like deals, assets or intentions.

But the tagging for each was managed by different editorial teams, who each used different terms.

I created a controlled vocabulary document to bring governance and a common language. Having defined consistent terminology, I considered the conceptual relevance of each term, and the part it played in helping real people complete real data research tasks.

For the proof of concept I was building, I promoted 'strategy', 'sector' and 'region' to taxonomy nodes. This allowed me to define the relationship each key object in the model had to the user's own idea of the tasks start-point.

It allowed queries like "Which investors said they would invest in one strategy, sector and region profile, but ended up investing elsewhere?" (strategy drift).