GDSN Is a Reality Check
GDSN is often presented as the solution to messy product information. I don't think that's quite true. The moment your data enters a serious standard, every shortcut suddenly appears under a spotlight: missing attributes, packaging fields nobody owns, inconsistent hierarchies, and supplier spreadsheets held together by copy-paste and optimism. The standard didn't create those problems. It simply stopped pretending they weren't there.
That becomes even more obvious when product data moves across borders. Because GS1 operates through a federated structure, the same data may need to pass local GDSN validation, target-market validation, and retailer-specific checks. Every layer makes sense on its own. Together, they create a surprisingly heavy process for something designed to simplify data exchange.
The Complexity Didn't Disappear
Large retailers can usually manage. They have the people, systems, and leverage to tell suppliers exactly how the data should look. Suppliers have fewer options. They inherit the mapping work, failed validations, corrections, new templates, and months of preparation. The complexity didn't disappear.
It simply changed desks.
That also creates a serious barrier for new suppliers. One route offers an intelligent transformation layer. The other asks suppliers to navigate a modelling stack before a product can even go live. Ten minutes versus ten months may sound dramatic. In some organisations, it isn't.
Which raises an uncomfortable question: if AI can already map, validate, enrich, and transform product data automatically, why are we still asking every supplier to rebuild the same data for every downstream system?
Why not transform it where the data actually enters the business?
The Amazing Part
None of this takes away from what GDSN gets right. A shared structure creates trust. Procurement, logistics, e-commerce, marketplaces, compliance, category management, analytics, and AI all benefit when they work from the same language. Product dimensions, weights, item numbers, ingredients, images, and packaging details become easier to reuse because everyone understands what they mean.
In theory, calm and elegant.
Commerce, unfortunately, did not get that memo.
Real product data arrives through Excel files, PDFs, APIs, emails, supplier portals, ERP exports, and whatever somebody generated five minutes before the deadline. Every supplier has different naming conventions. Every category has exceptions. Every retailer has slightly different requirements. Everyone is also convinced that their spreadsheet is the correct one.
GDSN doesn't remove that reality. It runs straight into it.
The Terrifying Part
The frightening part is not the standard itself. It's discovering how much manual work your business quietly depends on. Temporary spreadsheets somehow celebrate their fifth birthday. Mapping files only one person understands. Packaging data nobody owns. Validation rules living inside someone's head because they built the first import seven years ago and have accidentally become part of the infrastructure.
Everything works until the data has to leave the building.
That's usually when companies realise they don't have a GDSN problem. They have a transformation problem.
Why This Matters Now
The pressure on product data is increasing from every direction. The EU Green Deal, CSRD, CSDDD, ESPR, Digital Product Passports, packaging regulations, marketplace requirements, and sustainability reporting all demand more structured, transparent, and trustworthy information. AI doesn't make structured data less important either. It makes it more important. AI still needs context, consistent attributes, and information it can actually understand.
Garbage in, confident-looking garbage out.
GDSN Is a Checkpoint, Not the Destination
GDSN is often positioned as the answer. In reality, it standardises the conversation. It doesn't fix the data. It only reflects how ready, or unready, the data already is.
The future of product data exchange isn't just agreeing on one shared model. It's about what companies can do with their data before it reaches that model. AI-enabled mappings and transformations change the equation. Instead of forcing every supplier and system into rigid external constraints, businesses can introduce one intelligent layer that continuously cleans, validates, harmonises, and transforms product data.
The standard stays. The manual work doesn't have to.
Compared with that, GDSN starts to look less like the destination and more like a checkpoint. The real bottleneck isn't the standard. It's the fragmented, brittle transformation work happening before the data ever gets there: spreadsheets, mappings nobody owns, repeated corrections, and manual work that simply doesn't scale.
That's exactly where AI wins.
This blog post is a contribution by ProductFlight, written by Thorin Schiffer, Founder, CEO and CTO. ProductFlight helps organizations with AI-driven product data transformation for modern retail. For more information, visit www.productflight.io or stop by the ProductFlight booth at Data Expo.
