Product data enrichment, with a source for every value.

Supplier data is rarely ready for your shop

A typical supplier line looks like this: ‘CORDL. DRILL 18V 2X2AH CASE BLUE’. Abbreviations, no brand, no category. Before it can be listed, abbreviations must be expanded, attributes put into fields, the category found and a description written. For a few dozen items that works by hand; for thousands it does not. What is missing is not invented but marked as an open question.

What happens to a supplier line

Catalyst works in modules that you book one by one or run in sequence. The demo shows this with six fictional supplier lines; it is in German and uses prepared example outputs. The module pages are in German too.

• Normalisation: abbreviations, units and spellings are standardised first, for example ‘2X2AH’ becomes ‘2 batteries, 2.0 Ah each’.
• Categorisation: down to leaf level, in your own taxonomy or those of Google Shopping, Shopify and eBay.
• Attribute extraction: attributes from free text and data sheets as key-value pairs. What is not in the text stays empty.
• Attribute generation: missing attributes derived from category and context, marked as generated.
• Copy: product descriptions as text or HTML, bullet points and promotional copy.
• Marketplace attributes: required attributes for eBay, Amazon and Kaufland in each marketplace's format.

What each channel does with categories

• Shopify: Shopify says all products should be assigned a category from its Standard Product Taxonomy. The category enables category metafields (product attributes) and, with Shopify Tax, sets the tax rate at checkout.
• Google Merchant Center: Google assigns a product category automatically. The google_product_category attribute is optional and is meant for overriding that assignment in certain cases, such as campaign targeting.
• Your own shop or PIM: your category tree drives navigation and filters, so leaf-level accuracy matters most.

Three quality levels: Basic, Pro and Pro+

Pro+ checks against your raw data, not against the real product, so a final expert check by you is still recommended.

Prices per SKU and per category leaf

Prices are published on the pricing page (in German), per item (SKU) or, for taxonomy mapping and attribute enrichment, per category leaf. A monthly subscription suits a steady flow of new items; a batch job is a one-off order for existing stock at twice the subscription price per unit. Prices fall with volume in seven tiers from 50 to 10,000 units.

Extract for the 1,000-unit tier, subscription price per unit per month, in euros, net excluding VAT, for business customers only:

Worked example

1,000 new items from one supplier, as a one-off batch job in the 1,000-unit tier:

• Categorisation Pro: 1,000 × €0.36 = €360.00
• Attribute extraction Pro: 1,000 × €0.60 = €600.00
• Product description (HTML) Pro: 1,000 × €0.66 = €660.00
• Total: €1,620.00 net, or €1.62 per item.
• If about 200 new items then arrive each month, the same combination on subscription in the 200-unit tier costs 200 × (€0.30 + €0.50 + €0.55) = €270.00 net a month.

Formats and handover

• Input: CSV, XLSX or PDF data sheets as they come from the supplier, with no prior mapping to a fixed format.
• Output: CSV, XLSX or JSON, with category ID and path. The fields follow your data model.
• Handover: as a file for import, or directly to your shop system, PIM or ERP.
• Custom connection: if the handover needs a link to your stock system, we build it as an integration.

Enrichment or a text generator?

How an order works

1. 1Data review: you send an extract, and we recommend modules and levels.
2. 2Small run: an extract, from 50 units at list prices. You check it, and we set fields and thresholds.
3. 3Processing: the full range as a batch job, or new items on subscription.
4. 4Handover: result file or direct handover, plus the open questions for your supplier.