Back to updates
Private Preview
Insight Engine: analysis is meaningless before definitions settle
The most common blocker in preview is not model capability but the same metric meaning different things in different departments.
Natural-language querying looks like it lowers the barrier, but if "revenue" is computed three different ways across three systems, an accurate query is still meaningless.
So the first phase of every preview is settling metric definitions in one managed place, and only then opening natural-language questions. The generated query stays visible to the user so results can be checked.
Currently open to invited customers only.