Connecting product, customer, and marketing data at KEH

The work

At KEH, I worked on both sides of a retail data problem: making product information usable across systems and turning customer activity into better decisions.

I led an overhaul of the product information model for 100,000 SKUs, using data pipelines and integrations to bring information from multiple sources together. I also helped implement RudderStack data collection for more than 20 million monthly website events in BigQuery, then used segmentation to support marketing decisions.

In parallel, I built Python and Google Cloud models for price and inventory optimization using demand, seasonality, and price elasticity. I led the move from Universal Analytics to GA4 and developed dashboards and pipelines for marketing analytics.

The result

The product data became more useful for downstream systems, and the customer analysis gave marketing a clearer view of its audiences. I presented the findings to KEH’s CEO and marketing leadership. I also met one-on-one with the CEO for mentorship, gaining a direct view of the business priorities behind the data work.

Tools: Python, SQL, BigQuery, RudderStack, GA4, Qlik Sense, Google Cloud, GraphQL