The Resolution Layer · Case Study in BriefEKOM
Tires & batteries.
A cold read of the tire and battery lines of a national automotive service and tire retailer — where a shopper chooses a model in a size, and the model name, the page address and the specification table carry the answer alone. One of a set of anonymized EKOM analyses — what surfaced, and where it leads.
Lead finding · the same tire under two names
At least twenty-six tire models are filed under two spellings, splitting their sizes across two entries.
The spellings differ by a hyphen or slash, a capital letter or a stray double space. On one high-performance model about one size in eighteen sits under the second spelling, so a shopper who opens that entry sees a short list and may conclude their size is unavailable. Because tire pages are addressed by a slug built from the name, the split reaches the sitemap: eight tires are published at two addresses, and every page EKOM opened declares itself the original.
Also surfaced
A battery table filled by default
Ten battery pages read live carry the same placeholder values in the same fields: the cranking-amps field repeats the terminal type, and voltage and dimensions read "n/a" — even where the page address says 12V. The one figure that varies, cold cranking amps, is the one that is right.
A page with nothing behind it
A tire page with no record in site search dropped out of the sitemap but still serves, printing a JavaScript placeholder where its model name belongs.
A record with no commerce fields
Price, stock status and rating are empty on every record behind site search. Shoppers see prices at page render; anything reading the record instead sees a catalog with no prices and nothing in stock.
Codes and lost characters
Numeric category codes sit in a field meant for category names, and five "+" models are told apart from their predecessors only by a trailing hyphen in the address.
How EKOM read it. The record behind the retailer's site search holds little more than a name, a brand, a category, an image, a short description and a size count, so this read was a deterministic census rather than a model pass: every record in the index, with per-brand counts summing exactly to the index's own total, then the pages behind the findings opened in an ordinary browser. The count is low because the source is thin, not because the catalog is clean — and what is strong is strong: the sitemap publishes at least as many products as the index holds.
From diagnosis to channel-ready. The seven findings collapse into five upstream mechanisms — model names typed rather than governed, addresses generated from names, a specification template filling fields the product never supplied, internal values leaking into display fields, and two layers describing one catalog differently. Fixing the findings fixes the findings; fixing the mechanisms fixes every model already passed through them and every one that will. Then the same intelligence holds the line at intake, so the next model typed a second way never reaches a live page.