The Resolution Layer · Case Study in BriefEKOM
A national athletic-footwear retailer.
A footwear-catalog, no-schema read across sneakers, running, cleats, and kids' — one of a set of anonymized EKOM analyses. What surfaced, and where it leads.
Lead finding · discovery
In-stock shoes shoppers can't find.
Every size of a purchasable Jordan Luka 5 "Venom" is flagged non-searchable — in stock, correctly priced, invisible to on-site search. A Jordan 4 Retro "Toro Bravo" is tagged to the wrong Jordan retro collection, so it drops out of the collection filter shoppers browse. The highest-impact footwear failure isn't a missing product — it's a live, buyable shoe that never surfaces where a shopper looks.
Also surfaced
Discovery attributes blank
Collar height, running type/surface, cleat stud type, and kids' sizing tier run heavily blank — so a trail runner, a molded cleat, or a grade-school size never appears in the facet a buyer filters on.
Wrong or blank gender
An adidas F50 men's soccer cleat with a blank gender (invisible in gender nav); a women's shoe labeled "Men's"; nine Nike men's SKUs labeled "Women's Shoe."
The SKU layer
Non-searchable size runs, six Sabrina 3 "Silencer" variants pointing to the wrong parent product, price integrity (current > "original"; $0.00 items), blank/invalid UPCs, and zero shipping dimensions.
Missing fit & vendor fields
Missing sizeChartId (a top returns driver in footwear), and blank vendorName/vendorNumber on Jordan / Nike products despite a known brand.
How EKOM read it. With no schema and no manual setup, the engine profiled the catalog and detected that footwear attributes cluster by family (running, basketball, cleats, kids') — so it ran focused passes per family, reading sparse-but-family-defining attributes like cleat stud type, running surface, and collar height in the context where they matter, instead of averaging them away in a whole-catalog sweep.
From diagnosis to channel-ready. This pass read and diagnosed. The same understanding powers what follows — correcting the wrong-collection tags, blank/mislabeled genders, and broken variant parentage; enriching the footwear-specific discovery attributes plus sizeChartId so every shoe shows up on the filter a buyer uses; running the full variant pass; and holding the line at intake as new drops and colorways arrive.