joshgamache
commented
1 year ago See #179 for a continuation of this.
We're going to try and offload as much of the "Analytics" flow to Metabase as we can. Most of this can be shown in an acceptable form through that, reducing our front end dev load.
As a developer I want to implement the screens for the analytics user flow so that the "analyst" role can view all analytics data, view details of a dataset, perform calculation on a dataset, and export data
As an analyst, I need assurance that the data I'm working with in the analytics platform are from the data science workspace only, rather than from the data clean room,
Sprint 4 Direction Decision:
AW: I think for this proof of value we should be scrapping as many of the ideal state screens as possible. I'm suggesting that we absolutely should not build this in the scope of our proof of value. We should connect the (analytics) table to Metabase...and we should use a BI tool for showing the data that's in a table.
So in terms of in that context, limiting it down to what we need for the proof of value here, I think it is important as a key element of the story for our end users to understand that we have data going into this clean room that the Privacy concerns are being dealt with in the clean room, and that the analytics flow is on the analytics or data science workspace data where all the data we've put here is suitable for analysis.
Connecting a business intelligence tool and dumping the data analytics there is to fit into that broader story of bring your own BI tool and we're giving you some key insights in analytics that you can see in our system. But you can also see in your existing business intelligence stack.
Where this ticket fits in the broader vision:
Finding a technology that allows us to map our semantic layer of our data structure of our resulting models to various BI Tools. And therefore enabling us to connect the dots between the climate methodologies, which are very prescriptive, we have certain ways we need to calculate certain formulas, and there are consistent worldwide and the things we report out of those methodologies. While the methodologies may be in conflict with each other. The way we report tons of CO2 equivalent versus some economic indicator is the same regardless of our methodology that we've used to calculate the inventory to calculate the sum of CO2 E. Glue it together from the semantic layer to the business intelligence tool.
Acceptance Criteria:
Possible Future Requirements for Screen Mocks below:
Acceptance Criteria:
Wireframes
Hi Fi Mocks