Where they started
The brand did real volume, but 90%+ of it ran through the marketplace, and the own store was effectively invisible in search. The site had product pages and nothing above them: no category structure, no reason for a non-branded searcher to land there instead of on a marketplace listing. The most telling number was in the finance sheet, not the analytics: every marketplace sale gave up roughly 15% in fees, and the team had quietly accepted that as the cost of demand. People searching “best [product category] for [use case]” were being handed straight to competitors and to the marketplace’s own listings. The brand was paying to acquire customers it already had.
What we did
- Built category architecture around non-branded demand. We mapped the questions buyers ask before they know which brand they want, then built the category and sub-category pages that had never existed. That layer turns generic search into a reason to be on the site at all.
- Shipped comparison and buying-guide content. Honest “best X for Y” and “X vs Y” pages caught buyers mid-decision instead of post-decision, which is where the marketplace normally wins them.
- Captured branded search leaking to marketplace listings. For people already searching the brand by name, the own store had been losing the click to the marketplace listing; fixing titles, structure and internal links pulled that traffic back to the cheaper checkout.
- Earned 85 links from tech-review and community sources. The buying guides gave reviewers and forums something worth citing, which moved DR from 38 to 47 and let the new category pages actually rank.
None of this was a traffic play. It was a margin play: the same customer, routed through a checkout the brand controls instead of one that skims 15%. The curve compounded steadily because category authority builds slowly, and the revenue mix moved with it: 8% of DTC on the own store at the start, 31% by month twelve.