RegCatalog

Guide · Kaufland

Kaufland Product Data CSV: Format, Template & Required Fields

How to build a Kaufland product data CSV: semicolons, UTF-8, ean and locale, required fields, repeated picture columns, worked examples and a starter file.

By RegCatalogPublished Updated
On this page
  1. Quick answer: the Kaufland product data CSV format
  2. Product data is not offer data
  3. How the file is structured
  4. Field reference: the core columns
  5. Original CSV examples
  6. Starter file and official templates
  7. How to prepare and export the CSV in a spreadsheet
  8. Required attributes and why a valid file can still be rejected
  9. Validate the file before importing
  10. Frequently asked questions
  11. Sources

Last checked against Kaufland's Seller API documentation and Seller University: 23 October 2026

A Kaufland product data CSV is a plain-text file that describes products, one row per product and language, so that Kaufland can create or update its catalogue records. The format itself is short to describe. Most of the trouble comes from three places: confusing product data with offer data, not knowing which fields are required in which situation, and spreadsheet software quietly changing the file when it is saved.

This guide explains the documented file format, what each core column does, which fields are required and when, and how to produce the file from Excel, Google Sheets or LibreOffice without damaging it. All examples are original and synthetic. Nothing here is Kaufland's official template, and RegCatalog is not affiliated with Kaufland.

Quick answer: the Kaufland product data CSV format

Kaufland's Seller API documentation describes the product data file like this:

RuleWhat Kaufland documents
EncodingUTF-8
DelimiterSemicolon (;) between values
First lineA header row with the attribute names
Every following lineOne product
eanMust be present for every product; Kaufland identifies products by EAN
localeMust be present on every row, for example de-DE; one file may contain several locales
Several values for one attributeRepeat the column, one value per column (for example several picture columns)
categoryOptional in the file: it can be left blank or the column omitted (Seller University disagrees, see below)

If the file creates new products, Seller University adds the general mandatory attributes ean, category, title, description, manufacturer and picture, plus extra mandatory attributes in some categories.

A minimal file for one new product looks like this:

ean;locale;title;description;manufacturer;picture
2001234567893;de-DE;Akku-Bohrschrauber 18 V mit 2 Akkus und Koffer;Kompakter Bohrschrauber für Holz, Metall und Kunststoff mit zwei Gängen und LED-Arbeitslicht.;Example Manufacturer GmbH;https://images.example.com/2001234567893-1.jpg

Illustrative only, not for import. The EAN is a made-up value with a valid check digit. It uses GS1 prefix 200, which GS1 designates for restricted circulation numbers within a geographic region, not for globally unique trade item numbers; RegCatalog does not claim that the number is unused. The manufacturer is fictional, and the image address uses the reserved example.com domain.

Product data is not offer data

Kaufland keeps one shared product record per EAN. Every seller who sells that product attaches an offer to the same record. The Seller API calls these offers units. Product data says what the item is: EAN, title, description, manufacturer, images, category and attributes. The offer says how you sell it: price, stock and shipping.

That split decides whether you need a product data CSV at all:

Your situationWhat you sendProduct data CSV needed?
The EAN already exists in the Kaufland catalogueAn offer for the existing productNot to start selling. Seller University asks you to check that the existing product data is correct first.
The EAN does not exist yetNew product data, then an offerYes, with the general and any category-specific mandatory attributes
Existing product data is wrongCorrected product data for that EANYes. Seller University says you can pause the affected offers and upload corrected data without deleting the old data first.
Only price or stock changesOffer or inventory dataNo

A few consequences are easy to miss:

  • Adding an offer for an existing product is not the same as creating product data. If the EAN is already listed, a full new-product file is not the route to selling it.
  • Product data alone does not make a product visible. According to the Seller API, the product appears on the marketplace only once inventory (an offer) exists for it.
  • Your values may not all appear. The first upload for a product creates a description tied to your account, which Kaufland merges with data from other sources.
  • Product data cannot be deleted from the marketplace. You can remove your own contribution, but the public product description built from other sources remains.

Some third-party documentation that ranks for "Kaufland CSV required fields" describes offer feeds, with columns such as price, quantity and warehouse. Those are a different file. If your file has price and stock columns but no ean column with product descriptions, it is not a product data CSV.

How the file is structured

Encoding: UTF-8

Kaufland documents UTF-8. The quickest check is to open the file in a plain text editor and look at characters such as ä, ö, ü, é or ł. If they appear as ä or �, the text was converted with the wrong encoding somewhere along the way. Re-export it from the source system rather than repairing the characters by hand.

Kaufland's documentation says nothing about a byte order mark (BOM), the three invisible bytes that Excel's "CSV UTF-8" format writes at the start of a file. It is not proven to cause a failure, but removing it is the conservative choice.

Delimiter: a semicolon, even when a menu says "comma"

Values are separated by semicolons. A Kaufland help page for the product data file on kaufland.de tells sellers to save their spreadsheet with Excel's "CSV (Comma delimited)" option. That is the name of an Excel menu option, not the delimiter Kaufland expects. On Windows, Excel's CSV formats use the list separator from the regional settings. With German settings that is a semicolon. With UK or US settings it is a comma. Always check the saved file in a text editor.

The semicolon has one practical advantage: Kaufland also expects a decimal comma in numeric values (2,5), and commas inside values do not need quoting when the delimiter is a semicolon.

Header row and column names

The first line holds the attribute names, written exactly as Kaufland's technical names: ean, locale, title, not EAN or Product title.

  • Order: Kaufland identifies columns by their names and documents no fixed column order. Kaufland's own example file does not start with ean. Keep a consistent order anyway, because it makes files easier to compare.
  • Duplicates: Repeating a column name is how you send several values for one attribute. ean and locale must appear only once.
  • Your own attributes: Kaufland explicitly allows columns for attributes it does not know yet and encourages sending the data you have. For your own attributes, German or English names are both accepted.
  • Typos: A misspelled known name, such as manufactuer, is not rejected as an error. It becomes an unknown attribute, and the value never reaches the field you intended.

One row per product and locale

Every row carries a locale, the language and storefront that row's text is written for. Kaufland's documentation uses de-DE as its example and allows several locales in one file. When you supply text in more than one language, the same EAN appears once per locale. The same EAN twice with the same locale is a duplicate.

The text in each row must be in that row's language. Kaufland's product data guideline asks for correct language and rejects automatically generated translations in descriptions.

For storefronts other than Germany, take the exact locale codes from the Seller Portal or the Seller API documentation rather than guessing. Third-party integration documentation is not consistent. For Czech, for example, both the standard cs-CZ and the non-standard cz-CZ appear.

Repeated columns for several values

Kaufland does not want several values packed into one cell with a separator. Instead, the column is repeated. A product with three images has three picture columns.

For images with alternative text, Kaufland documents column pairs: picture.url holds the image link and picture.alt_text the description of that image. Kaufland's product data guideline says alt text has been a legal requirement since June 2025, and that alt text without a matching image link causes an error.

Empty fields

An empty value is simply nothing between two semicolons. Each row must still have the same number of fields as the header, so a product with fewer images than there are picture columns ends with empty fields such as ;;.

Empty fields are fine for optional values. The Seller API explicitly allows category to be left blank. In a row that creates a new product, empty mandatory fields make the product data invalid.

Kaufland's file documentation does not say whether an empty cell removes a value you sent earlier. Do not rely on empty cells to delete data.

Quoting values

Kaufland's file documentation does not describe quoting rules. Spreadsheet exporters and most CSV tools follow the common convention (RFC 4180):

  • a value that contains a semicolon, a double quote or a line break is wrapped in double quotes;
  • a double quote inside such a value is written twice ("");
  • other values need no quotes, although quoting them is harmless.

A missing closing quote is one of the most damaging structural errors. The parser keeps reading until it finds the next quote, which can merge several products into one value.

Value formats inside cells

The Seller API's product data page defines how values of each attribute type are written. The ones you are most likely to meet:

Value typeDocumented formatExample
Decimal numberDecimal comma; a period only as a thousands separator2,5
Number with unitNumber followed by a unit250mm
DateDay.month.year5.3.2026
Yes/notrue/false, 1/0, T/F, yes/no, Y/N or ja/neinyes
PicturePublic URL of a jpg, png or gif filehttps://images.example.com/1.jpg
EANA valid EAN2001234567893

Which type applies to which attribute is defined in Kaufland's attribute list, which RegCatalog does not reproduce.

Field reference: the core columns

The table distinguishes what the documentation requires from what it recommends. "New products" means rows that create product data for an EAN that is not yet in the catalogue.

ColumnWhat it holdsStatusPractical notes
eanThe product's EAN (GTIN)Required on every rowMust be a valid EAN with a correct check digit. Keep it as text. One EAN per product: Kaufland's guideline treats listing one product under several EANs as spam.
localeThe language and storefront of the rowRequired on every rowFor example de-DE. Use the codes Kaufland gives for your storefronts.
categoryThe Kaufland categorySources conflict (see below)A leaf category or a higher-level category that Kaufland narrows down. The category decides which category-specific attributes apply.
titleProduct nameMandatory for new productsThe guideline recommends 50 to 80 characters, with no promotional text, symbols or seller information.
descriptionProduct descriptionMandatory for new productsThe guideline recommends 150 to 5,000 characters. Plain text or HTML with lists and bold headings; no links, images, videos or JavaScript.
manufacturerManufacturer or brand nameMandatory for new productsIf Kaufland does not know the manufacturer yet, Seller University asks you to send the brand name through the contact form.
picturePublic image URL; repeat the column for more imagesMandatory for new products (at least one)The Seller API names jpg, png and gif; the guideline also lists WebP and TIFF. The guideline recommends 5 to 10 images, at least 1024 px and at most 10 MB.
picture.url and picture.alt_textImage URL and its alternative text, as adjacent pairsUse when you supply alt textAlt text without a matching picture.url causes an error.
videoPublic direct link to an MP4 fileOptionalMay be left empty.
is_smart_device and smart_device_manufacturer_information_urlEU Data Act status and manufacturer information URLFor connected productsSee the Kaufland EU Data Act seller guide.
Category-specific attributesFor example the material composition of textilesMandatory in the affected categoriesListed in Kaufland's per-category overview.
Your own additional attributesAny further product factsOptionalAllowed and encouraged by Kaufland; names in German or English.

It helps to read the requirements as layers:

  1. File-level requirements apply to every file, including updates: UTF-8, semicolons, a header row, and ean and locale on every row.
  2. Mandatory attributes apply when product data is created. Without the general mandatory attributes, product data counts as invalid and is not displayed. Some categories add their own mandatory attributes.
  3. Recommendations improve quality but are not import rules: title and description lengths, image count and resolution.

The category discrepancy

Kaufland's sources disagree on whether category must be provided:

  • Seller University (mandatory attributes): lists category as one of the general mandatory attributes for all categories.
  • Seller API (product data CSV files): describes category as optional. You can leave it blank for some or all products, or omit the column. Kaufland then selects a suitable category.

One way to read both is that every product needs a category, but Kaufland can assign one when the file does not contain it. Kaufland does not state this, so treat it as unresolved. If you know the correct category, sending it gives you more control. If you leave it out, check how Kaufland categorised the product, because the category determines which category-specific attributes become mandatory. RegCatalog's checker reports a missing category as information, never as an error.

Original CSV examples

The examples below are synthetic. EANs are made-up values from GS1 prefix range 200–299, which GS1 designates for restricted circulation within a geographic region; they are not presented as unused or as numbers to list on Kaufland. The manufacturer is fictional, and image addresses use example.com. Descriptions are shortened to keep lines readable and are shorter than Kaufland's recommended minimum. None of these files is ready for marketplace import.

Example 1: empty fields and repeated picture columns

ean;locale;category;title;description;manufacturer;picture;picture;picture
2001234567893;de-DE;;Akku-Bohrschrauber 18 V mit 2 Akkus und Koffer;Kompakter Bohrschrauber für Holz, Metall und Kunststoff. Zwei Gänge, LED-Licht, Ladegerät im Lieferumfang.;Example Manufacturer GmbH;https://images.example.com/2001234567893-1.jpg;https://images.example.com/2001234567893-2.jpg;https://images.example.com/2001234567893-3.jpg
2009876543213;de-DE;;Gartenschlauch 20 m mit Schnellkupplung;Flexibler Gartenschlauch mit Gewebeeinlage, knickfest, mit Schnellkupplung und Spritzdüse.;Example Manufacturer GmbH;https://images.example.com/2009876543213-1.jpg;;

What to notice:

  • category is empty in both rows (;; after the locale), which the Seller API allows.
  • The commas inside titles and descriptions need no quotes, because the delimiter is a semicolon.
  • The second product has one image. The two unused picture fields stay as empty values at the end of the row, so both rows have nine fields.

Example 2: images with alternative text

ean;locale;title;picture.url;picture.alt_text;picture.url;picture.alt_text
2005550001116;de-DE;Schreibtischlampe 40 cm, schwenkbar, Metall;https://images.example.com/2005550001116-1.jpg;Schwarze Schreibtischlampe mit schwenkbarem Arm, Frontansicht;https://images.example.com/2005550001116-2.jpg;Lampenkopf mit Drehgelenk in Nahaufnahme

Each picture.url is directly followed by its picture.alt_text. This example shows only the columns that matter here; a new product would also need description and manufacturer. The next row is wrong: the second image link is missing but its alt text is still there, which Kaufland reports as an error.

2005550001116;de-DE;Schreibtischlampe 40 cm, schwenkbar, Metall;https://images.example.com/2005550001116-1.jpg;Schwarze Schreibtischlampe mit schwenkbarem Arm, Frontansicht;;Lampenkopf mit Drehgelenk in Nahaufnahme

Example 3: quoting and formatted descriptions

ean;locale;title;description
2003141592654;de-DE;"Kabelbinder 200 Stück; schwarz";"Kabelbinder aus Polyamid, UV-beständig. Farbton ""tiefschwarz"", Länge 200 mm."

The title contains a semicolon, so it is quoted. The description contains double quotes, so it is quoted and each inner quote is doubled. Without the quotes, the title would split into two fields and every later value in the row would shift one column to the right.

If you want structure in a description, Kaufland's guideline allows HTML with lists. That avoids raw line breaks inside a value, which are valid CSV when quoted but are easy to damage in other tools:

ean;locale;title;description
2003141592654;de-DE;Kabelbinder 200 Stück, schwarz;<p>Kabelbinder aus Polyamid, UV-beständig.</p><ul><li>Länge 200 mm</li><li>Breite 4,8 mm</li></ul>

Example 4: what spreadsheet damage looks like

ean;locale;title
2,00123E+12;de-DE;Akku-Bohrschrauber 18 V mit 2 Akkus und Koffer
200123456788;de-DE;Werkzeugkoffer leer, 45 x 33 x 15 cm
2009876543218;de-DE;Gartenschlauch 20 m mit Schnellkupplung
  • Row 1: the EAN was saved in scientific notation. The original digits are gone from this file and cannot be recalculated. Re-export them from the source system.
  • Row 2: the original value was 0200123456788. The spreadsheet dropped the leading zero. The remaining 12 digits are still a valid GTIN-12, because a leading zero does not change the check digit, so no checker can tell from the file alone that a digit is missing. Compare the value with the packaging or the source system.
  • Row 3: the check digit is wrong (the correct EAN would end in 3). Typical causes are a typo, swapped digits or a value from the wrong column.

Example 5: a comma-delimited export

ean,locale,title,manufacturer
2001234567893,de-DE,"Akku-Bohrschrauber 18 V, mit Koffer",Example Manufacturer GmbH

This is a valid CSV file, but not in Kaufland's format: Kaufland would read each line as a single column. Do not fix it with search and replace, because that also turns the comma inside the quoted title into a delimiter. Re-save the file with a semicolon delimiter in a tool that understands CSV quoting, or use a converter that does.

Starter file and official templates

Kaufland provides its own templates. Seller University's instructions for CSV uploads start with downloading the category-specific template, and the Seller API page on product data files links an example file with 100 products. Use Kaufland's category template when you build files for a specific category, because it reflects the attributes that category uses.

How to prepare and export the CSV in a spreadsheet

The safest file is one exported directly from your ERP or PIM system and never opened in a spreadsheet. When a spreadsheet is unavoidable, the main rule is that the EAN column must be text before any data enters it. Once a spreadsheet has converted an EAN to a number, the damage is usually already done.

Never "fix" EANs by padding them with zeros automatically. A 12-digit value may be a valid GTIN-12 or an EAN-13 that lost its leading zero, and only the packaging or the source system can tell you which.

Microsoft Excel (Windows)

  1. Do not open the CSV by double-clicking it. Use Data › From Text/CSV.
  2. In the import dialog, set File Origin to 65001: Unicode (UTF-8), the Delimiter to Semicolon, and Data Type Detection to Do not detect data types. Every column then stays text, including EANs.
  3. For a new sheet, format the EAN column as Text before you paste or type values.
  4. Recent Microsoft 365 versions also have File › Options › Data › Automatic Data Conversion, where you can stop Excel from removing leading zeros and from using scientific notation for long numbers.
  5. Save with File › Save As › CSV UTF-8 (Comma delimited). Excel writes the list separator from your Windows regional settings: a semicolon with German settings, a comma with UK or US settings. It also adds a BOM.
  6. Open the saved file in a text editor and check the delimiter, special characters and EANs.

Google Sheets

  1. Use File › Import, choose a custom separator ;, and untick Convert text to numbers, dates, and formulas.
  2. Format the EAN column as Format › Number › Plain text.
  3. File › Download › Comma-separated values always writes commas. Convert the delimiter afterwards with a CSV-aware tool (see example 5), never with search and replace.

LibreOffice Calc

  1. When opening, choose Unicode (UTF-8) as the character set and Semicolon as the separator. Select the EAN column in the preview and set its column type to Text.
  2. Save with File › Save As, file type Text CSV, and tick Edit filter settings. Choose UTF-8, ; as the field delimiter and " as the string delimiter.

Check the file before you upload it

Open the exported file in a plain text editor and work through this list:

  1. The first line is the header, with exact names and exactly one ean and one locale column.
  2. Values are separated by semicolons.
  3. Special characters such as ä, é or ł look correct.
  4. Every EAN is complete: digits only, no spaces, no E+. Compare 12-digit values with the packaging.
  5. Every row has the same number of fields as the header.
  6. Rows that create new products have a title, description, manufacturer and at least one image.
  7. Image URLs open the image itself in a private browser window, not a login page or an HTML page.
  8. Your categories' extra mandatory attributes are present.
  9. Values starting with =, +, - or @ are intended and were not created by a spreadsheet formula.

Required attributes and why a valid file can still be rejected

General requirements for new products

Seller University names six general mandatory attributes for every category: EAN, category, title, description, manufacturer and picture. Product data missing mandatory attributes is invalid and is not displayed. Such products appear in the Seller Portal's "Invalid items" report.

Category-specific requirements

Some categories add mandatory attributes. Seller University's examples include the material composition of textiles, ingredients and nutrition information for food, the energy efficiency class, energy label and EU product data sheet for large appliances, TVs and monitors, age ratings for films and games, safety guidance for toys, the signal word for hazardous chemicals and the fuel efficiency class of car tyres. For some of these categories, Seller University says the item cannot be offered for sale without them.

The complete list per category is published by Kaufland as an overview file. RegCatalog does not reproduce it, and its checker does not test category completeness.

Why a structurally valid CSV may still be rejected

A file can have perfect structure and still fail, because Kaufland checks the content too:

  • a category-specific mandatory attribute is missing;
  • a value is not on Kaufland's list of accepted values for that attribute;
  • the manufacturer is not yet known to Kaufland;
  • image URLs are not public, not image files, or not reachable;
  • text is in the wrong language for the row's locale, or machine-translated;
  • the product duplicates an existing listing under another EAN;
  • the category management team reviews some values manually, which according to the Seller API can take hours or days.

There are also documented limits: a CSV file imported from a URL may be at most 1.5 GB, and the Seller API caps product feed imports at 30 per day, so it is better to send many products in one file.

After an import, the Seller Portal's product data reports show what happened: the import error report, invalid items, EANs not found, and an export for improving existing product data. If an import has already failed, our troubleshooting guide maps common symptoms to their causes and fixes: Kaufland-CSV-Import: Fehler beheben (in German).

Validate the file before importing

RegCatalog's free Kaufland CSV Checker checks a product data CSV before you upload it. It runs entirely in your browser: the file is processed locally and is neither uploaded nor stored.

What it checks:

  • File and encoding: UTF-8, BOM, UTF-16, double-encoded characters, file size up to 50 MB.
  • Delimiter and structure: semicolons, quoting, the number of fields per row, empty lines.
  • Header row: the ean and locale columns, duplicate and empty column names, wrong capitalisation of core names.
  • EANs: length, check digit, scientific notation, duplicates within a locale, and a note on 12-digit values.
  • Locale: the format of each locale value.
  • Documented core fields: title, description, manufacturer and image URLs (as warnings, because they are mandatory mainly for new products), image and alt-text pairing, and a missing category as information.
  • Selected formatting issues: surrounding whitespace, formula-like values and the two EU Data Act fields.

What it does not check:

  • accepted values of attributes that have value lists;
  • names and formats of attributes outside the core fields, which are shown as "not checked", not as errors;
  • category-specific mandatory attributes;
  • whether a locale belongs to a Kaufland storefront;
  • offer files and Seller Portal reports;
  • what is behind an image URL, such as resolution or reachability.

The checker can propose safe corrections, such as converting the delimiter, converting to UTF-8, removing a BOM, fixing core column names and trimming whitespace. It applies only the corrections you approve. EAN digits are never changed, and missing leading zeros are never added.

A result without errors does not mean Kaufland will accept the file. It means the checks that were performed found nothing. The import log in the Seller Portal remains the final word.

Frequently asked questions

Is there an official Kaufland product data CSV template?

Yes. Seller University's CSV upload instructions start with downloading a category-specific template, and the Seller API documentation links an example file. The starter file on this page is RegCatalog's own header-only file with the core columns: an unofficial preparation aid, not an official or import-ready Kaufland template.

Why does a Kaufland help page mention "CSV (Comma delimited)" when the file needs semicolons?

"CSV (Comma delimited)" is the name of an Excel save option. On Windows, Excel uses the list separator from the regional settings, which is a semicolon with German settings. The Seller API documentation is explicit that values are separated by semicolons. Check the saved file, not the menu name.

Do I need a product data file if my EAN already exists on Kaufland?

Not to start selling it. You create an offer for the existing product. Seller University asks you to check that the existing product data is correct first. Upload product data only to add missing information or correct errors.

Do I have to include the category column?

The sources disagree. Seller University lists category as a general mandatory attribute. The Seller API file documentation allows blank values or no column, in which case Kaufland chooses the category. If you know the right category, sending it gives you more control.

Can one CSV contain several languages?

Yes. Every row has its own locale, and Kaufland allows several locales per file. Repeat the EAN once per locale, write each row's text in that row's language, and avoid two rows with the same EAN and locale.

Can I add my own columns?

Yes. Kaufland accepts attributes it does not know yet and encourages sending the data you have, with names in German or English. Make sure Kaufland's own attribute names are spelled exactly, because a misspelled known name is treated as a new attribute.

Does the column order matter?

Kaufland documents no fixed order and identifies columns by name. Keep each picture.url directly before its picture.alt_text, and keep ean and locale to one column each.

Can I prepare the file in Google Sheets?

Yes, but set the EAN column to plain text before entering data, and note that the CSV download is always comma-delimited. Convert the delimiter to semicolons with a CSV-aware tool before you upload.

Will a file that passes RegCatalog's checker be accepted by Kaufland?

Not necessarily. The checker tests structure, encoding, EANs and documented core fields. It does not check category-specific mandatory attributes or accepted values, and Kaufland may review content manually.

Sources

RegCatalog is an independent service and is not affiliated with, endorsed by or reviewed by Kaufland. The examples on this page are synthetic and do not reproduce Kaufland's templates, example files or attribute lists. Kaufland's requirements can change; the linked original sources and the import log in the Seller Portal are authoritative. This guide is not legal advice.