Case study: B2B locksmith / key supply on BigCommerce — search with fitment in mind
Anonymized B2B BigCommerce catalog: keys, locks, and vehicle fitment — where “close enough” search results waste a locksmith’s day.
B2B key and locksmith supply catalogs are dense: restricted lines, dealer-only SKUs, and shoppers who search like technicians — partial keyway codes, machine models, and “the blank we used last time.” On BigCommerce, native search often treats that vocabulary as noise.
This anonymized case study is about a B2B BigCommerce storefront where Seekmodo search had to respect who can see what, while still helping authorized buyers find the right blank or tool on the first try. No merchant name or logo — pattern only.
The search problem in a B2B catalog
- Codes and abbreviations that never appear in the public product name
- Customer-group visibility — results must not leak restricted products to the wrong segment
- Repeat buyers who remember last year’s SKU better than this year’s title
- Mobile punch-in from a van where typing precision is low
Native BigCommerce search is serviceable for tidy retail titles. It struggles when the “query” is a partial industry code and the stakes are a locksmith’s time on-site.
What we needed from Seekmodo
Install via the BigCommerce App Marketplace path, keep Stencil chrome intact, and improve typeahead + SERP relevance without standing up a separate Elastic project. Enhanced Native had to be acceptable on day one; cloud ranking and analytics came after pairing.
Vehicle-style fitment is not always the metaphor — here the analog is compatibility codes and machine models. The retrieval problem is the same: map messy human language onto structured catalog fields.
What we installed
- Seekmodo BigCommerce app / connector with storefront search swap and suggest
- Index rules aligned with visibility so unauthorized SKUs do not surface
- Cloud ranking for click learning on technician queries
- Admin pinning for critical blanks and tools that must win navigational searches
MCP mattered later for internal tools — dealer ops asking an agent “do we stock this blank?” against the same index buyers search. That is optional; the storefront win came first.
Outcomes we will stand behind qualitatively
Authorized buyers found common blanks faster via typeahead. Support fewer “send me the link” requests for SKUs that should have been searchable. Staff trusted search enough to use it on the phone with customers instead of only browsing categories.
Named revenue lifts stay off this page without written approval. The operating question was: does search reduce friction for people who already know what they want?
If you run B2B on BigCommerce
Audit twenty real technician queries before you buy any search product. If native search fails half of them, you have a retrieval problem, not a training problem. Seekmodo’s BigCommerce path is documented at /solutions/bigcommerce-search-app and /plugins/bigcommerce; vehicle-oriented catalogs may also care about /solutions/bigcommerce-vehicle-search.
B2B-specific gotchas
Index and search against the customer group the shopper actually belongs to — otherwise you will “fix” relevance by showing products they cannot buy. Pin the navigational blanks that drive phone orders. Treat partial code queries as first-class test cases in QA, not afterthoughts.
If you also run a public marketing site on WordPress beside the B2B catalog, keep tenants separate; shared relevance is fine, shared indexes usually are not.
Next step
Install from the marketplace listing described in our BigCommerce launch post, start Enhanced Native, then trial cloud ranking from /pricing. Compare search-volume billing philosophy on /compare/doofinder if finance is weighing keystroke-metered alternatives.