The 4.9% of visitors who searched brought 27.5% of revenue.
A European kids' fashion and accessories store switched on personalized search. Parents search in their own words, and results are ranked for each shopper.
- Kids' fashion and accessories · EU
- 35 days, mid-2026
- Searchers against non-searchers, same window
Parents search by age, size and occasion. Product names do not.
Kids' fashion is searched in parent language: “sandals girl 28” or “wedding outfit boy”. A default store search matches keywords against product names, so it returns too little, too much or the wrong things.
When search fails, the shopper digs through category pages or leaves, and the store never learns what they wanted.
Search that knows the shopper
Suggestions from the first keystroke
Ranked by the live catalog and by each shopper's own behavior: what they browsed, and the sizes and categories they return to.
One memory behind it
Search runs on the same profile as recommendations, email and chat, so every query sharpens the next message too.
Demand the store can read
Every query is intent, including for products the catalog is missing.
Shoppers ran 118,000 full searches in the same period.
Searchers against everyone else
Same 35-day window, so promotions affect both groups. Searchers chose to search, so this is not a controlled test.
| Measure | Did not search | Used search | Share or change |
|---|---|---|---|
| Visitors | 497,126 | 25,806 | 4.9% of visitors |
| Buyer rate | 0.59% | 2.64% | 4.5× |
| Revenue per visitor, indexed | 100 | 724 | 7.2× |
| Purchases | 3,514 | 923 | 21% of purchases |
How we counted
Self-selected. Searchers are the store's highest-intent visitors, so part of the gap is intent, not search.
What it shows. Revenue concentrates in search. Every improvement to it works on more than a quarter of the store's revenue. The same store runs the chat on the same memory, for shoppers whose question does not fit a search box.
Get the method and setup notes
How each number on this page was counted, the definitions, and what was switched on in the store, in order. Enter your work email and the notes open; we also send you the link.
Thanks. Open the method and setup notes. The link is on its way to your inbox too.
We email you the link and add you to Releva's emails on ecommerce research and news. We also see what you read on this site, and our team may get in touch. Unsubscribe any time. Privacy policy
Read next
The chat on the same store
Buyers the agent advises spend two to nearly three times more per order.
Read the case study →SolutionHelp shoppers find the right product
Search, pages and chat that answer in the shopper's own words.
See the solution →ProductSales agent
Search, recommendations, chat and banners on your site, decided for each visitor.
See the Sales agent →
See what your best visitors are typing
One call, with your store open. We will show you your top searches and what they return today.
See it on your store
You'll talk to one of us.