ETIM classification for Pimcore products
Export your Pimcore product objects to CSV or Excel and get the same file back with the ETIM class and features of each product, checked against the ETIM model.
How the Pimcore export is read
The Pimcore export is recognised by the “id”, “fullpath” and “published” columns. In Pimcore each installation defines its own data classes, so the rest of the header is yours.
Pimcore's technical columns (“id”, “fullpath”, “published”, creation and modification dates) are left out. For the others a role is suggested from the column name, in Spanish or English: reference, name, description, brand, GTIN or attribute. Check it in the second step: that is where it shows most that every Pimcore is different.
A reference column is required (SKU, article code…). Rows without a reference are not processed.
From the Pimcore export to ETIM in four steps
- 01
Upload the file
CSV, Excel .xlsx or legacy .xls. The file is read in your browser: it is never uploaded whole to any server.
- 02
Check the columns
A role is suggested for each column (reference, name, description, brand, GTIN, category, attribute) and you correct it.
- 03
Choose the ETIM version and run
Available versions: ETIM 7.0, 8.0, 9.0 and 10.0. Each product gets a class and the values of its features.
- 04
Review and download
Correct whatever needs correcting on screen and download as Excel, BMEcat ETIM 5.0 or ETIM xChange 2.0.
What you get
- The same file you uploaded, with the ETIM class of each product, its features and an “Unmapped” sheet with everything that did not fit the model.
- Also BMEcat ETIM 5.0 and ETIM xChange 2.0, the two exchange formats that distributors and wholesalers ask for.
- The codes (EC, EF, EV) are the official ones. Names are in ETIM English, the only open language of the model; your products can be in any language.
What is checked before you get the result
- The class must exist in the chosen ETIM version. If none fits, the product is left with “no class” and the reason; it is not given the closest match.
- Each feature only accepts what its class allows: a value from its list, a number, a range or yes/no. Everything else goes to “Unmapped”.
- Units are converted with a table, not by the language model. An unknown unit, or one that measures a different quantity, is not converted.
- Each value carries a literal quote from the product data. If the quote is not in your file, the value is discarded.
Price and limits
One credit classifies one product. When you create your account and verify your email you get 20 trial credits. Trial accounts accept up to 200 products per file.
Frequently asked questions
Do my products have to be in English?
No. They can be in Spanish or any other language. Classification searches the ETIM catalogue in English and returns the official codes, which do not depend on language.
Which ETIM version is used for classification?
The one you choose: ETIM 7.0, 8.0, 9.0 and 10.0. A saved result can later be converted to another version without classifying again. ETIM 11.0 will be available as soon as ETIM International publishes it (expected on 1 December 2026).
Can I load the result into my PIM without reviewing it?
It is not advisable. The codes exist and the values are allowed by the class, but the choice is proposed by a language model and can be wrong. Review the low-confidence rows, the “no class” rows and the “Unmapped” sheet first.
What happens to my product data?
The file is read in your browser. The fields you select are sent in batches to the Claude API, from Anthropic, to propose the class and the values. The reference, name and result of each product are kept for 30 days so that you can download it again.
Have another question? Write to us