Keyword Search Is Dying. Is Your Catalog Ready for AI Agents? — PayPal
- Source
- youtube.com
- Author
- AI Engineer
- Date
Context
Giving shopping agents more product information helped in PayPal's reported catalog experiments, but adding text indiscriminately could backfire. The conference session's description says enriching merchant product records consistently improved recommendations, and that catalogs with thin information gained the most. Over-enriching with unstructured boilerplate diluted the useful signal and led agents to make up details.
The implication for merchants is narrow: catalog preparation should improve the information an agent uses to match a product to a shopper's request, not simply lengthen descriptions. The description reports these findings qualitatively and gives no effect sizes or sales results, and the talk itself was not reviewed.
- PayPal's Nixon Dinh describes shopping moving from a search era of typed keywords, to an intent era where shoppers describe needs, to a delegation era where agents do the searching.
- He argues catalogs syndicated for ads and keyword SEO are not readable by agents, and that hybrid keyword plus semantic search works better than either approach alone.
- In PayPal's experiments, enriching merchant product data consistently improved agent recommendations, and thin catalogs gained the most.
- The limit he reports: over-enriching with unstructured boilerplate diluted the signal and made agents , so content quality mattered more than adding text.
- AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
- hybrid search — Combining keyword and semantic search in one ranked result, so exact terms and paraphrases both find what they should.
- hallucination — When a model states something false with full confidence — inventing facts, citations, or APIs that don't exist.
Checking sign-in…
Loading comments…






