Marketplace Attribute Requirements: A PIM Playbook

Marketplace attribute requirements are no longer a once-a-year feed-cleanup task. In 2026, Amazon, Google Merchant Center, bol.com, Zalando, Kaufland and TikTok Shop all treat structured product attributes as eligibility signals: if the right value is missing, badly mapped or placed in the wrong field, the listing may never publish, may lose visibility, or may be pushed into an exception queue just before a campaign goes live.

The uncomfortable part for multichannel sellers is that each marketplace defines “complete” differently. Google expects a stable core of attributes such as id, title, description, link, image, availability and price, then adds conditional requirements for GTIN, brand, apparel variants, energy labels and more. Amazon’s product type rules can require hundreds of extra fields across categories. bol.com exposes conditional attributes in its Product Content API. Zalando separates general article data from mandatory, mandatory-if-applicable and category-specific material rules. Kaufland’s API shows required attributes by category, including EAN, manufacturer, title and category for many flows. A spreadsheet can store this data, but it rarely governs it.

That is why the PIM question has changed. The best product information management setup is not just a nice product catalogue with images and translations. For marketplace sellers, it must work like an operational control layer: one canonical product record, channel-specific mappings, validation rules before export, and a clear owner for every exception. ChannelDock’s PIM feeds and PIM feature overview are built around that operating model: enrich once, adapt per channel, and stop bad data before it becomes a rejected feed.

Why attribute requirements are now an operations problem

Most ranking content treats attribute requirements as a static checklist: add GTIN, fill in brand, add colour and size, submit the feed. That advice is useful for a single channel, but it breaks when a seller lists the same SKU on Amazon, bol.com, Zalando, Kaufland, Shopify, Google Shopping and TikTok Shop. The same product attribute may need a different field name, accepted value, language, unit, format, category mapping or “not applicable” treatment per marketplace.

For example, a shoe listing is not just “material: leather”. Zalando’s article mapping guidance asks for separate material specifications such as upper material, lining or inner material, insole material and sole material. Google may care about colour, size, gender and age group for apparel visibility. Kaufland’s category endpoint returns category-specific required and optional attributes. TikTok Shop’s Partner Center exposes built-in product and sales attributes by leaf category, with mandatory attributes depending on listing policy. The operational work is not writing a description; it is maintaining a living ruleset.

The four-layer PIM model for marketplace attributes

A practical PIM for marketplace attribute requirements needs four layers. The first is the canonical product record: brand, SKU, GTIN or EAN, manufacturer, measurements, weight, material, safety data, media, translations and internal taxonomy. This is the source that your team trusts. It should not be polluted with every channel’s field names.

The second layer is channel mapping. This turns canonical fields into marketplace-specific destinations: colour to color, EAN to external product ID, material composition split into multiple footwear fields, and “one size” converted into the accepted value for the target channel. The third layer is validation. Before export, the PIM should check required fields, accepted values, character limits, unit formats, image requirements, identifiers, category dependencies and conditional attributes. The fourth layer is release control: which products are ready, which are blocked, who owns the missing value, and whether the block is a revenue risk.

What current PIM guides usually miss

Competitor pages from Akeneo, Plytix, Pimcore, Salsify and Inriver explain the value of centralised product data, attribute mapping, data quality and channel syndication well. The gap is that many guides stop at software capability. They say “map attributes” or “keep up with retailer requirements”, but they rarely show how a lean seller team should run the weekly process when requirements change, when a marketplace rejects a subset of SKUs, or when supplier data arrives incomplete.

Forum threads show the operational pain more clearly. Shopify merchants report Google Merchant Center feeds where attributes disappear, titles do not update, variant fields are missing or products are disapproved after syncing through an app. Amazon sellers complain about listings suppressed for a missing Unit Count attribute even after values were entered, GTIN exemption confusion, and required fields that feel irrelevant to the product. These are not copywriting problems. They are governance problems: unclear ownership, no test export, no before/after diff, and no way to separate one-off fixes from durable PIM rules.

The winning PIM workflow is not “fill every field”. It is “know which fields become mandatory for this product, on this channel, in this category, before the marketplace tells you.”

A weekly attribute governance workflow

The right rhythm is simple enough for a small ecommerce team and strict enough for a 10,000-SKU catalogue. Run it every week, and always before peak campaigns, new marketplace launches or large supplier imports.

1. 1
   Pull the latest marketplace requirements

   Review seller documentation, API schema changes and rejection reports for Amazon, bol.com, Zalando, Kaufland, Google and TikTok Shop. Store changes as rules, not as notes in someone’s inbox.
2. 2
   Classify SKUs by risk

   Segment by revenue, stock position, campaign priority, new marketplace launch and category complexity. A missing attribute on a top-selling SKU matters more than the same error on a dormant item.
3. 3
   Validate before feed export

   Run category, identifier, media, language, character-limit and conditional-attribute checks inside the PIM. Do not wait for marketplace rejection files to become the first test.
4. 4
   Fix the rule, not only the SKU

   If fifty apparel variants are missing age group, add a category rule. If only two products need manual material values, assign the owner and deadline. Separate system fixes from data-entry fixes.
5. 5
   Measure accepted listings after publish

   Track accepted, rejected, suppressed, limited-performance and overwritten SKUs by channel. Feed success is not “file uploaded”; it is “sellable listing visible with correct content”.

How to design rules without creating bad data

Automation can make product data better, but it can also multiply a wrong assumption across thousands of listings. The safest rule design uses three levels. Auto-fill values only when the input is deterministic: country of origin from supplier master data, default language from market, package dimensions from the ERP, or “unisex” when the assortment policy says the product is intentionally unisex. Suggest values when the input needs human judgement: material, occasion, use case, compatibility, age group, sustainability claim. Block export when the value would be risky to guess: safety warnings, CE labels, EPREL certification, GTIN exemptions, regulated claims, hazmat data and product compliance documents.

This is where integrations matter. A PIM cannot validate marketplace requirements if it is detached from the webshop, ERP, WMS, supplier files and order system. Product data and operational data meet in the feed: availability must match stock, dimensions affect shipping rules, bundle definitions affect identifiers, and locale-specific content affects category visibility. Treating PIM as a marketing-only database creates beautiful product pages that still fail at marketplace submission.

Marketplace-specific checks to include

For Amazon, check product type, variation theme, GTIN or exemption status, brand authority, unit count, condition, images, title length and any newly required category fields before creating or editing listings. For Google Merchant Center, check the core required attributes, price and availability consistency, identifiers, apparel attributes, item group ID, image rules and landing-page match. For bol.com, treat conditional attributes as first-class logic: one selected value may create a new requirement elsewhere in the product record.

For Zalando, build separate controls for material composition, category-specific article data, warnings, measurements and mandatory-if-applicable attributes. The March 2025 Zalando changes reported by marketplace tooling vendors — with optional attributes removed, some becoming mandatory and new attributes introduced — are a good reminder that fashion and lifestyle data requirements move quickly. For Kaufland, read category-specific required attributes from the API and keep language, category, EAN, manufacturer and title controls visible to the team. For TikTok Shop, make sure the leaf category is correct before mapping attributes, because mandatory product and sales attributes are category dependent.

What to measure after the rules go live

A marketplace attribute project should not be judged by “number of fields completed”. That rewards busywork and encourages teams to fill weak or irrelevant values. Better KPIs connect data quality to sellability and speed.

• Pre-export pass rate: percentage of SKUs that pass PIM validation before feed submission.
• Accepted listing rate: percentage of submitted SKUs accepted without rejection, suppression or limited-performance warnings.
• Time to publish: hours from product-ready in ERP or Shopify to sellable on the marketplace.
• Repeat-error rate: how often the same missing attribute appears after it was supposedly fixed.
• Revenue at risk: value of stock or sales velocity attached to blocked SKUs.

FAQ

Conclusion

Marketplace attribute requirements will keep changing because marketplaces use structured data for compliance, search, filters, ads, AI shopping experiences and customer confidence. Sellers who manage those requirements in spreadsheets will keep reacting to rejection reports. Sellers who turn them into PIM rules can launch channels faster, reduce suppressed listings and keep product data consistent across every marketplace.

The practical next step is to audit one high-revenue category across three channels. List the required, conditional and recommended attributes. Map them to your canonical product fields. Add validation before export. Then measure accepted listing rate and repeat-error rate for the next two feed cycles. That small exercise usually shows whether the business has a product data problem — or a product data governance problem.