From Cross-Sell to Shopper Intent: What Experimentation Taught Us

Walmart Connect’s experimentation and analytics team sought to expand sponsored-product coverage on grocery search pages, where advertising inventory is sparse, initially testing complementary product ads such as chocolate or smoothie ingredients alongside banana searches. Although shoppers engaged with these cross-sell recommendations, the experiment did not reliably increase purchases, leading the team to conclude that broadly complementary items could distract from a shopper’s immediate goal. It redesigned the approach around inferred shopping missions, grouping a smaller set of relevant products into themes such as Taco Tuesday, which improved click-through and conversion outcomes. Because its tests run on pages owned by teams with different priorities, Walmart Connect coordinates objectives, establishes acceptable declines in page-owner metrics, and monitors revenue, relevance, and latency, including long-running ad-free holdouts to assess advertising’s effect on customer trust. The team emphasizes experiments as tools for learning rather than solely launch decisions, using predefined hypotheses, test plans, safeguards against early result-checking, and standardized end-of-test reviews, while expanding self-service analysis tools so product managers and others can conduct more experiments.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.