Retail media scorecard across nine retail networks
Rebuilt an agency’s HTML scorecard as an 8-page Power BI product covering Walmart Connect, Kroger, Instacart, Amazon Ads and more.
34,000+
automated checks, zero mismatches vs. source
- Client
- CPG marketing agency
- Location
- USA
The problem
The agency reported spend, revenue and ROAS for food and beverage brands from a spreadsheet and a static web page. Period logic was inconsistent and every client review started with reconciling numbers.
What I built
- 01
Consolidated history and the live year into one append-only fact table so future years load without rework.
- 02
Designed 40 KPI cards with month-over-month and year-over-year comparisons and color-coded movement.
- 03
Built a watchlist page that flags brands under the ROAS floor and the revenue gap to target.
- 04
Wrote a Python validation harness that tied every number back to the client’s own workbook.
The result
- Client reviews start with decisions, not reconciliation.
- Underperforming brands surface automatically each month.
- Every KPI is proven against source across 50+ filter scenarios.