Aglet

Learn from product search intent drift for the next review cycle

Search intent lessons are useful when they preserve the original query and the evidence that changed its classification. Compare the first assumption with later clicks, reformulations, and support outcomes. Turn the difference into one review question rather than a permanent rule about every ambiguous phrase.

Keep the lesson for the next incident

  1. Keep the original signal

    Record the entry point, phrase, result classes, and first interpretation that shaped the decision. Note what was not known at the time. This protects a later review from treating an eventual click as if it had been obvious initially.

  2. Compare later journeys

    Revisit related queries after the experience changes or more evidence arrives. Look for repeated shifts between shopping, support, and navigation. Preserve counterexamples because they show whether the lesson is a query-family boundary or only a single phrase.

  3. Change one review question

    Adopt one durable prompt, such as asking what result type would complete the job or checking the entry surface before reading click data. Assign an owner and date for trying it, then define the observation that would show it helped.

What to carry forward

The learning record should connect one intent classification to later evidence and one review adjustment. Stop when another reviewer can apply the question to a new query without inheriting an unsupported rule. Carry forward uncertainty where the journey still cannot separate intent families.

Technical background: Digital.gov guidance.

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