Analytics · Data model
No warehouse required to start
Eazy Analytics runs against the same operational database the platform writes to, which means reporting is available on day one rather than after a data-warehouse project.
Why this is the default
Most mid-size manufacturers do not have a data team. A reporting layer that requires a warehouse, a pipeline and somebody to own both is a reporting layer that is still six months away when the question is asked.
Eazy Analytics reads the operational records directly. There is no extract, no nightly load and no window during which the report and the system disagree.
What that costs
It is an honest trade, not a free win:
- Heavy historical analysis competes with transactional load. In practice this becomes noticeable in the tens of millions of documents, not before.
- Retention is whatever the operational database keeps. A warehouse is where multi-year trend analysis belongs.
- Combining Eazy data with a third-party source needs somewhere to combine it.
When to move
The model is dimensional from the start — conformed customer, product, territory and time — so moving it into a warehouse is a load job rather than a redesign. That is the whole point of doing the modelling work up front on a system that does not yet need it.