Guest writer, Cresco managing director Mark A. Gray, rethinks product enrichment to meet the demands of modern fashion shoppers.
Being a part of online e-commerce now for the last 20+ years, online fashion has become a passion of mine. Here, I share some ideas and considerations for fashion retailers and designers as we go into the busiest sales season in ANZ.
Across ANZ, fashion retailers continue to face one of the most deceptively complex challenges in e-commerce – i.e., managing product data at scale. As consumers’ expectations rise and marketplace algorithms evolve, the quality of product information has shifted from a back-end operational concern to a direct driver of sales performance. The shift towards automation, better enrichment, and AI-supported classification is no longer a future ambition. It is becoming a benchmark for retailers who want to compete in a market defined by speed, precision, and constant discovery.
The ANZ fashion landscape is unique in that retailers often sell across multiple channels, owned e-commerce, domestic and international marketplaces, dropship partners, and emerging social commerce platforms. Each channel behaves differently, yet all rely on clean, structured, consistent product data.
This is where many fashion brands, even well-known ones, still struggle. Key attributes such as size, fabric composition, fit, colour, and style are often not standardised. Variant relationships can be unclear, and category structures differ across platforms. Operational teams must navigate internal systems not designed for marketplace selling while managing the pressure of updating hundreds or thousands of SKUs each season.
One of the most overlooked areas is taxonomy. Every marketplace uses its own category structure, often requiring far more detailed paths than older ERP or POS systems support. Without automated taxonomy mapping, merchandising teams spend countless hours manually assigning products to the right category on each channel. This is inefficient and increases the risk of misclassification, affecting ranking and discoverability. Automation makes the process faster and more accurate, ensuring each item lands where customers and algorithms expect it.
Product enrichment is another area where manual processes fall behind. Retailers using spreadsheets and human input eventually hit a ceiling: attributes become inconsistent, titles drift, and descriptions lose the precision needed for search. Automation helps standardise this, but the real opportunity is AI-supported enrichment. Recent AI advancements can detect missing attributes, recommend improvements, align descriptions to marketplace requirements, and optimise titles for ranking without compromising brand tone or accuracy.
The difficulty becomes more evident as apparel moves into peak trading. When volumes surge from late October through early January, small issues in product data become magnified. If size curves are incomplete, if colours are not mapped accurately, or if attributes are missing, products fail to surface in the search filter; they lose visibility to competitors, resulting in traffic declines and conversion dips. And because marketplace algorithms prioritise listing quality, poor content can push entire product ranges down the ranking ladder at the very moment retailers need maximum exposure.
When data is incomplete or inconsistent, it doesn’t just impact discoverability but also returns. Fashion already carries one of the highest return rates in e-commerce, and ANZ shoppers are particularly sensitive to accuracy in size, material feel and product imagery. If key attributes are missing or unclear, customers are more likely to buy the wrong size or the wrong variation, leading to costly returns, cancellations, and abandoned baskets. In a high-cost operating environment, every preventable return directly affects margins. Intelligent product data management becomes both a commercial and operational advantage.
The next shift in data enrichment will come from AI-powered content generation and automated taxonomy detection. As retailers expand to more channels, including fast-growing Southeast Asian marketplaces, the need for high-quality, structured product data will intensify. Systems will need to make intelligent decisions—spotting gaps, generating missing attributes, and aligning content to each platform’s ranking logic. Early adopters are already seeing faster speed-to-market, stronger consistency, and clear conversion uplift.
Retailers who succeed will be those who treat product data as a strategic asset, not a back-office task. Clean, automated, intelligently managed data is now central to online fashion performance in ANZ. It drives discovery, reduces returns, boosts marketplace ranking, and shapes the customer experience. With peak season nearing, there has never been a more important time to get it right.
CrescoData is a cloud-based technology business serving enterprise customers globally. Our CrescoData proprietary platform was designed with scalability to keep up with the rapidly growing Commerce demand. https://crescodata.com
This is a partnered content series with eStar.