Aglet

Learn from zero-result recovery and empty-state decisions

Zero-result lessons should preserve why a search ended empty and which recovery respected the original job. Compare later queries with the original matrix, then improve one review question about filters, vocabulary, scope, or truthful availability instead of treating every empty result alike.

Keep the lesson for the next incident

  1. Keep the empty boundary

    Record query, scope, filters, result type, empty copy, near matches, and recovery action. Note whether the result was unavailable or only narrowed away. This protects later reviews from calling a valid empty state a retrieval failure.

  2. Compare later searches

    Review new hard and filtered zeros after the change. Look for a second filter combination without recovery or a suggestion that broadens too far. Preserve a working empty state and a populated control to keep the lesson bounded.

  3. Add one empty-state check

    Require future reviews to classify the zero, preserve the original phrase, test one removal or wording path, and state when no alternative exists. Assign ownership and revisit date. Define the observation that would show the check reduced dead ends.

What to carry forward

The learning record connects an empty state to a tested recovery or an honest unavailable message and one reusable review check. Stop when another reviewer can classify hard and filtered zeros. Carry forward uncertainty where catalog or index evidence cannot explain the absence.

Technical background: W3C 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