An EU automotive parts retailer ran a clean A/B test: 1,300 visitors, 6,400 searches split between default PrestaShop search and BRADsearch. CTR +42%, conversion rate +63%, revenue per search +124%.
This customer is an EU specialist in automotive exterior styling, shipping across the EU and to the USA. Their catalogue covers front bumpers, rear diffusers, side skirts, spoilers, and German vehicle styling products - designed in the EU and manufactured to specification. (Shared anonymized: this store ran an A/B test but is referenced without its name.)
Their buyers are different from general retail shoppers. They arrive at the site knowing exactly what they need: a specific part, for a specific car model, often in a specific style or material. The search bar is not a discovery tool - it is the fastest route from intent to purchase. When it fails, the buyer leaves.
PrestaShop default search returned poorly ranked results for automotive queries. Buyers who searched for "F30 front bumper" or "BMW E46 diffuser" would encounter a mix of loosely related products sorted by keyword overlap rather than relevance. Scrolling through irrelevant results is friction - and in automotive parts retail, friction means the buyer goes to a competitor who shows them the right part immediately.
Automotive buyers use chassis codes - F30, F82, E46, G20. They add style variants: M-sport, M-performance, carbon, gloss black. They combine these with part types and brand references into structured queries that default search cannot parse. "F30 M-sport front bumper carbon" is not four keywords; it is a specification. PrestaShop's default search treated it as four keywords and matched accordingly.
Without search analytics, there was no visibility into which vehicle models generated demand but returned no results - a direct signal for which product lines to expand. Demand was invisible, which meant catalogue decisions were made without the most relevant data point available.
For a product category where visual differences are commercially significant - the difference between an M-sport bumper and a standard bumper, between gloss and matte, between styles - showing product images during autocomplete changes the dynamic entirely. Buyers can see the part before they click. Autocomplete now surfaces product thumbnails alongside category and brand suggestions, reducing misclicks and increasing confidence in the result.
BRADsearch handles chassis codes as meaningful identifiers, not arbitrary letter-number strings. Multi-attribute queries - brand plus vehicle model plus part type plus style specification - are parsed and matched against the full attribute set of each product, not just the product title. Typos in chassis codes and model names are resolved with typo tolerance tuned for automotive terminology. These are structured queries, and BRADsearch treats them as such.
Search analytics now reveal which vehicle models and part types generate search queries that return no results. For a specialist retailer, this is a direct product assortment intelligence feed: customers are telling you exactly what they want to buy, and analytics make that signal visible and actionable.
The A/B test ran 6,400 searches across 1,300 visitors - a clean, controlled comparison between identical traffic split between default PrestaShop search and BRADsearch. Every metric moved significantly in one direction.
For a store where buyers arrive knowing exactly what they want, these improvements compound directly into overall revenue uplift. A buyer who searches, clicks, and converts in one session needs no further persuasion - they needed only a search bar that understood them.
← Back to all case studiesJoin merchants across Europe who use BRADsearch to turn every search into a sale.
✓ No credit card required · ✓ Free onboarding setup · ✓ Cancel any time