AI System for Grouping Product Variants into the Correct Family

The problem

In day-to-day operations, SKU, brand, model, colour and size fields arrive in separate records and files, preventing the team from reviewing the same case in one place.

As a result, the link to the source can be lost while the draft product family is prepared, leading to repeated checks and uncertainty about the basis for the decision.

Incomplete automation may overlook similar-looking items that are different models, while poorly controlled automation creates a risk of confusion in stock and pricing records.

What we set out to improve

• Time taken for the draft product family to be ready for review
• Accurate routing of similar-looking items that may be different models to the correct specialist queue
• Recording of suggestions amended or rejected by the catalogue manager

How the system works

1. 01 Retrieve SKU, brand, model, colour and size data from an approved source
2. 02 Verify the source system, record identifier and freshness information
3. 03 Transform the fields into the target process schema
4. 04 Prepare a draft product family
5. 05 Compare the draft with business rules and source records
6. 06 Exception check: similar-looking items may be different models
7. 07 Catalogue manager approval, amendment or rejection
8. 08 Write the input, suggestion, amendment and final decision to the audit log

Methods we used

• Source-traceable data extraction for SKU, brand, model, colour and size fields
• Structured output that restricts the product-family fields through business rules
• An exception gate that separates similar-looking items that may be different models
• A queue that shows the source and suggestion together for the catalogue manager
• Duplicate-processing and rollback records to manage the risk of confusion in stock and pricing records

Where people stay involved

Data and security

Only the fields required for the task are processed from the SKU, brand, model, colour and size data. Source-system permissions are preserved, and personal and commercially sensitive data is minimised before being passed to the model. Every read, suggestion, human amendment and write to the target system is logged.

Who this suits

Frequently asked questions