Cleaning Product Data for a Fashion E-commerce Catalogue Migration
A fashion e-commerce retailer moving to a new platform needed structured product attributes for its filters, while the old catalogue held most of them as free text. Product data cleansing with agreed attribute mapping and validation rules made the catalogue ready before go-live.
Product Data Challenge Before the E-commerce Catalogue Migration
The new e-commerce platform needed structured values for colour, material, fit and category to build its filters. In the old catalogue most of these were free text, written differently by different people.
The migration date was fixed, so the data had to be cleaned in weeks, not months.
- Free-text colours and materials
- Products with missing categories
- A fixed migration deadline
What Was Done to Clean the Catalogue Product Data Before Migration
Attribute Mapping from the Old to the New Catalogue
Every old value mapped to the new attribute structure.
Validation Rules Before Catalogue Import
Blocking checks that stopped incomplete products before import.
Exception List for Manual Product Data Review
The remaining cases, each with an owner and a status.
Catalogue Product Data Cleansing Results in Numbers
attributes mapped
SKUs blocked at go-live
Tools Used in the Catalogue Product Data Project
Power Query Attribute Mapping
Repeatable remapping of attribute values.
Excel Exception Lists
Owner and status for every open case.
Related Service: Product Data Quality and PLM Standardisation
Attribute dictionaries, mapping and validation rules for product data in PLM and e-commerce systems.
Filters only work if the data under them agrees.
Related Retail Analytics Articles
PIM vs PLM for Fashion Brands: What Each System Does
PLM manages a garment from design to production; PIM prepares its data for selling. What each system holds, who owns it and when you need both.
Product Attribute Standards for Fashion PLM That Work Across Markets
Which product attributes a fashion brand needs in PLM, and how a shared dictionary keeps colour, season and composition consistent in every market.
Catalogue Product Data Cleansing Case Study FAQ
The project took five weeks, fitted around a fixed migration date. Attribute mapping came first, then the validation rules, and the remaining cases were handled through an exception list.
A full export of the old catalogue, including free-text colours, materials, fits and categories, and the attribute structure required by the new e-commerce platform. The platform’s filter requirements defined which fields had to be complete.
The team moved to the new platform with structured attributes, so the site filters worked from day one. Validation rules now stop incomplete products before import, and open cases are tracked with an owner and a status.
Retail and Fashion Analytics Case Studies
Standardising Product Data Across European and Asian Markets in a PLM System
A fashion brand selling in Europe and Asia replaced local naming habits with one PLM attribute dictionary and validation at entry.
Rebuilding the Womenswear Range Plan for a European Fashion Distributor
Fewer options, more depth in proven sizes and colours, and a new open-to-buy plan for a womenswear distributor.
Building a Price Architecture and Margin Report for a Fashion E-commerce Retailer
Clear entry, core and top price points per category, one set of pricing rules for all markets and a weekly margin report.