Aglet

Investigate a product search synonym gap with paired queries

An investigation must show whether two expressions retrieve the same useful product set under the same context. Replay literal and alternate terms, vary category or locale only when evidence calls for it, and record both rescued searches and newly introduced false matches.

Build a useful investigation brief

  1. Build the term matrix

    List literal terms, alternate terms, language or region, category, and expected product family. Capture returned titles, facets, and first useful action for each query. Include terms that should not match so the mapping has a negative control.

  2. Probe the boundary

    Test the candidate mapping in the narrowest relevant scope, then one neighboring scope. Compare overlap, missing variants, and unrelated results. Record whether a correction changes the submitted phrase or only broadens retrieval; those behaviors have different user expectations.

  3. Package the finding

    State the equivalent meaning, supported scope, false-positive cases, and remaining hypotheses. Link representative queries and explain which observation would justify expansion or rollback. Keep a content or indexing cause separate from the vocabulary conclusion.

What to carry forward

The investigation is complete when paired terms have a reproducible overlap pattern and known exclusions. Stop with a scoped mapping or an explicit inability to prove equivalence. A successful rescue does not outweigh a severe false match without documenting the tradeoff.

Technical background: GOV.UK developer documentation.

Keep the decision with the work.

Use a Work Item in Aglet to record the problem, the evidence you have, and the next decision. Add an owner and priority, then keep updates in the discussion so the next person can follow the reasoning.

Create an account See the product workflow