Enterprise commerce retailer
Technical search optimization and large-scale data enrichment
Reverse engineered search behavior and rebuilt the data layer that paid and organic channels ran on, producing step-change improvements in cost-to-acquire and organic reach.
600%
Paid-search ROI improvement
3x
Customer acquisition
10x
Organic traffic growth
Business challenge
The search channel (organic and paid) was operating with both hands tied. Product data was missing the attributes that ranking and bidding actually weighted, and the team treated SEO and PPC as marketing functions instead of search-engineering problems. Spend was wasted on queries that were never going to convert, and the catalog wasn't visible for the queries that would.
Approach
Treated search behavior as a system to model, not a channel to buy. Reverse engineered which query and product attributes were driving ranking, then enriched the catalog at scale to surface those attributes consistently. Pipelines pulled from multiple structured and unstructured sources to fill product-data gaps the merchandising team didn't have time to chase manually.
On the paid side, PPC management was automated against the same enriched dataset, so bids were tied to product economics and intent rather than blanket category rules. Google Shopping submissions were rebuilt around the enriched feed so the catalog finally showed up correctly in the surfaces that mattered.
Impact
Paid-search ROI improved roughly 600%. Customer acquisition tripled. Organic traffic grew about 10x. The unlock in every case was the same: putting useful data in front of the algorithms customers were ultimately searching through.