← Work
2021

Customer segmentation with BigQuery + Looker Studio

A SQL-first segmentation that runs entirely in the warehouse and lands in a dashboard. No separate ML stack required.

Context

Most segmentation tutorials assume a Python notebook, a data scientist, and a way to operationalise the output. Many teams have none of that, but they do have BigQuery and a BI tool.

Approach

  • RFM-style scoring written as a single BigQuery query against transactional data.
  • Segment definitions surfaced as a Looker Studio dashboard, so marketing can filter and act without re-running anything.
  • Designed for teams where the data scientist isn't always in the room.

Outcome

A reproducible, warehouse-native segmentation pattern, published as a step-by-step guide on Towards Data Science.

Stack

  • BigQuery
  • SQL
  • Looker Studio
  • RFM

Links