E-commerce & Digital

Product categorisation at scale

Automatically assign products to the correct site taxonomy categories using title, description, and attributes. Removes a bottleneck on new product onboarding.

AutomationPattern-matchingAccuracyTime saving

80–90%

reduction in categorisation time

Opportunity assessment

Business Impact
2

Minor improvement. Small efficiency gain with limited effect on overall turnover or bottom line.

Feasibility
5

Very easy to implement. Ready-to-use tools exist. Can be up and running in days or weeks with minimal technical resource.

Data Readiness
4

Light data requirements. Uses straightforward inputs — documents, product descriptions, customer records — that are usually accessible with minimal prep.

Risk Exposure
4

Low risk. Limited external exposure. A human reviews output before it reaches anyone outside the team.

Change Complexity
5

Very easy to absorb. Runs quietly in the background or gives people a helpful new input. Makes everyday work easier with almost no friction.

Tooling required

Standard LLMWorkflow automation

Things to consider

  • Walmart used predictive and generative AI to improve product categorisation across its catalogue, enabling AI-powered use-case search rather than individual item search (Walmart Global Tech).

  • Taxonomy clarity is the prerequisite — ambiguous or overlapping categories confuse the AI just as they confuse humans.

  • Exceptions handling: define a threshold below which the AI routes to human review rather than auto-assigning (e.g. confidence below X%).

  • Monitor accuracy monthly and retrain prompts when new categories are added or naming conventions change.

  • Misclassification has a direct impact on findability — products in the wrong category don't appear in the right search results or navigation paths. Prioritise accuracy over speed during calibration.

  • Experiment starter: Take 200 recently onboarded products that were manually categorised. Run the same products through a prompted LLM against your taxonomy and measure accuracy by category. If overall accuracy exceeds 90% with no individual category below 80%, the case for automation is made.

AI Transformation Playbook

Ready to assess your own opportunities?

The playbook gives you the full 150-opportunity directory, scoring tools, and 100+ templates for every stage of an AI transformation programme.