Identity came before volume
Every imported record was checked for a stable source identifier before publication. Duplicate IDs were rejected, one record with a missing title remained in Draft, and product URLs were built around the numeric catalog identity rather than a brand label. That means an editorial correction can improve a title or category without silently creating a second product page.
The source fields remain separate from editorial fields. Importing a title and image does not authorize the catalog to invent a price, rating, material or availability statement. Those unknowns stay unknown until current information is provided for the exact item.
Categories support discovery without redefining products
Explicit source-category mappings connected products to twelve primary category hubs and a set of maintained secondary routes. The mapping rules are recorded rather than inferred from free-form titles. Secondary pages must now meet a twenty-product threshold before they can enter the index, which keeps small or empty combinations available for internal filtering without presenting them as full landing pages.
What visitors can do now
Visitors can search an exact ID, browse a category, compare real image galleries, save products locally and send a shortlist through WhatsApp. The site still has no checkout and does not display a synthetic price. The next useful catalog improvement will come from better evidence and clearer navigation, not from changing the identity of the 21,380 published records.
Useful notes
- Use Product IDs when returning to a saved item.
- Treat category counts as catalog counts, not stock.
- Report a mismatch with the exact item URL.