Data · 2019 – now
Pharma commercial analytics platform
100+ production Power BI dashboards for six global pharma companies, on models past 500 million rows, each serving up to 1,000 people.
500M+
rows in one model
- 100+
- dashboards in production
- 500M+
- rows in the largest model
- 1,000
- users on one report
Problem
US commercial teams at Sanofi, GSK, AstraZeneca, Gilead, CSL and ViiV needed one trusted view of omnichannel marketing, patient journeys, NBRx/TRx, HCP reach, payer and rebate performance. The data lived in Snowflake, SQL Server and SSAS Tabular at a scale where a naive model refreshed for hours and opened in tens of seconds.
Approach
Star-schema semantic models with aggregation tables and incremental refresh; measures instead of calculated columns; high-cardinality columns removed; bidirectional relationships eliminated; query folding enforced in Power Query. Workspaces, deployment pipelines, RBAC and row-level security designed to pass pharma compliance audits.
The craft
Performance work is invisible when it succeeds, so it was measured: DAX Studio and Tabular Editor runs before and after every model change, with refresh windows and report load times tracked per release.
Result
Refresh windows went from hours to minutes and report loads stayed in seconds at enterprise scale. A 12-person team ships and maintains the estate; Copilot for Power BI was integrated for natural-language questions and auto-written narratives, with enablement sessions run as an Axtria-certified trainer.
Stack
- Power BI
- Snowflake
- SQL Server
- SSAS Tabular
- DAX Studio
- Tabular Editor
- Power Automate