Multi-Channel Analytics for a DTC Nutrition Brand

The situation

A direct-to-consumer nutrition brand sells across three distinct channels: its own Shopify storefront, Amazon, and a subscription program managed through Recharge. Each channel generates its own data in its own format. Executive reporting required manually reconciling numbers pulled from three sources, and the team’s director of IT was standardizing on Microsoft infrastructure, which meant migrating from Looker Studio to Power BI.

Marketing attribution compounded the problem. The brand spends across Google, Meta, and Amazon Ads, and the marketing team had no reliable way to see how those dollars traced back to revenue by channel or by SKU. When a new product flavor launched, adding it to reports was a manual process. When fraud occurred, there was no dedicated dashboard to investigate it.

The company had data flowing in but no coherent infrastructure to make it useful at the pace the business required.

What we built

CorrDyn stood up a data engineering foundation using Estuary to replace a Fivetran-based ELT stack, starting with the Shopify connector and extending to Recharge as the engagement continued. Estuary’s pricing model fit the brand’s data volume better than Fivetran’s row-based billing, and the migration reduced pipeline costs while maintaining the same data freshness.

On top of the pipeline layer, we built a dbt transformation layer that standardized metric definitions across channels and modeled SKU-level performance data in a form PowerBI could consume efficiently. We migrated the executive reporting suite from Looker Studio to PowerBI, building dashboards for executive reporting, operational metrics, and fraud monitoring.

For marketing analytics, we automated the ingestion of channel spend data from Northbeam, connecting it to revenue data from Shopify and Amazon so the team could see blended efficiency metrics without manual assembly. Ship-by date reporting and data oddity investigations across dashboards became part of regular operating rhythm. The engagement runs on a monthly retainer with a Principal, Senior Data Analyst, Data Analyst, and Project Manager covering new integrations and reporting requests as the business evolves.

What changed

The executive team now pulls reports from a single PowerBI environment instead of reconciling numbers across three tools. When the brand added new SKUs, they were modeled and tracked inside the existing analytics infrastructure rather than patched into ad hoc spreadsheets. Marketing decisions about channel spend now draw from a unified attribution view instead of platform-native dashboards that each report in isolation. The fraud dashboard gave the operations team a structured way to investigate irregular activity. The pipeline cost reduction freed budget that went back into analytics development.