Consolidated nine disconnected data sources across five retail brands into a single Snowflake data warehouse and self-service BI platform — cutting monthly reporting time from 12 days to same-day.
Bramwell Retail Group operates five distinct retail brands, each running its own POS, eCommerce, and inventory systems. Every month-end, a team of three analysts spent up to twelve days manually exporting and reconciling spreadsheets from nine different systems just to produce a single consolidated revenue report — by which point the numbers were already three weeks stale.
Digivance designed and built a unified Snowflake data warehouse with dbt-managed transformation models and Fivetran-based ingestion pipelines from all nine source systems. On top of that foundation, we delivered a self-service Power BI layer so brand managers can now answer their own questions instantly instead of filing a report request and waiting weeks.
Five brands each ran independent POS, eCommerce, and inventory systems with no shared data model, making any cross-brand analysis a manual, error-prone spreadsheet exercise.
Three analysts spent up to twelve days every month manually exporting, cleaning, and reconciling data from nine systems just to produce one consolidated revenue report.
Different brands and teams routinely reported different numbers for the same metric, since each was pulling from a different system with a different definition of "revenue" or "active customer".
By the time the monthly report was finished, the data was already three to four weeks old — too late to inform any real-time merchandising or inventory decisions.
Any ad-hoc data request required a specific analyst to manually pull and format the data, since there was no self-service layer non-technical stakeholders could safely use themselves.
Several source systems had inconsistent or missing historical records, making year-over-year trend analysis unreliable without substantial manual cleanup each time.
We designed a Snowflake-based data warehouse as the new single source of truth, with a dimensional model built around consistent, agreed-upon definitions for revenue, customers, and inventory across all five brands.
Fivetran connectors were configured for all nine source systems, replacing manual CSV exports with automated, reliable data ingestion running on a schedule with built-in monitoring and alerting.
Business logic and metric definitions were codified in version-controlled dbt models, so "revenue" and "active customer" mean the same thing everywhere in the organisation, tested automatically on every change.
Data pipeline scheduling and dependency management runs through Airflow, giving the team full visibility into pipeline health and immediate alerting when a source system's data doesn't arrive as expected.
Built role-specific Power BI dashboards on top of the governed warehouse, letting brand managers and executives answer their own questions in real time instead of waiting weeks for an analyst to run a report.
Established clear data ownership, access controls, and a documented data catalogue so the warehouse stays trustworthy and navigable as more data sources and users are added over time.
Snowflake warehouse architecture and dbt transformation framework were live within 8 weeks, with the first three of nine source systems fully integrated and validated against manual reconciliation.
All nine source systems fully integrated into the unified warehouse. Self-service Power BI dashboards launched to brand managers and executive leadership across all five brands.
Monthly close process reduced from 12 days to same-day. Three analysts previously dedicated to manual reconciliation reassigned to higher-value analysis work instead.
"We used to dread month-end — twelve days of spreadsheet archaeology across five brands that never quite agreed with each other. Now our brand managers pull up a dashboard and get the answer in seconds, and for the first time, everyone is looking at the same numbers. It's changed how fast we can actually make decisions."
Let's talk about building a data foundation your whole team can actually use.