Use cases · Catalogue content

How it works

You keep one master product record; the model reshapes it per channel — Amazon's title conventions and five bullet points, bol.com's Dutch-language titles without prices, promotions or special characters, eBay's item specifics — following a written rule sheet you maintain for each marketplace. It only rewrites and reformats facts you supply; it is not asked to know anything about your products or the marketplaces on its own.

Data you need

Your master product content, plus a current, written style and rules document per marketplace and category: character limits, forbidden content, required attributes. Most shops do not have such a rule sheet written down yet — expect a one-off effort to distil it from each marketplace's seller documentation before this works. The model's built-in knowledge of marketplace rules is stale by definition; the rule sheets must come from you and be kept current.

What to expect

Removes the grind of maintaining three or four versions of every listing by hand, and applies your rules consistently once they are written down. Compliance rests on your rule sheets, not on the model — marketplace category requirements change often and no offline model tracks them. Even with good rules the model can occasionally drop a required attribute, run over a character limit, or slip in promotional wording a marketplace forbids, so listing rejection remains possible. Treat first batches per category as tests. If you have no written rule sheets, building them is the real first step of this project.

Where people stay involved

Someone who runs the marketplace account reviews output per category before upload and owns keeping the rule sheets up to date when marketplaces change policy. For bol.com, have a Dutch speaker spot-check the first batches — the model writes competent Dutch, but nuance and category jargon deserve a native eye.

• Turn supplier spec sheets into first-draft product descriptions

  Drafts product descriptions overnight from the supplier data you already hold, in your shop's voice — a review job instead of a from-scratch writing job.

• Pull sizes, materials, colours and other attributes out of messy supplier text

  Reads free-text supplier info and fills a fixed field list — material, dimensions, colour, EAN — returning structured rows for your PIM or shop admin.

• Sort new products into the right category in your shop

  Embeddings shortlist the most plausible categories for each new product; a small model picks one and explains why — workable even on large category trees.

• Write page titles and meta descriptions for every product and category page

  Drafts an SEO title and meta description for every page from real product attributes, within your character limits — economical even at long-tail scale.