Aglet

How to prioritize slow search by task value and reach

Prioritizing slow search starts with what the search enables. Consider whether people are looking for a primary record, how many terms or filters reproduce the delay, whether every keystroke triggers work, and whether the behavior appeared after a release or data growth.

Decide where the work belongs

  1. Name the discovery work that is blocked

    Describe the next action after results arrive, such as opening a work item or finding a customer record. Distinguish an annoying delay from a search that prevents the task from completing. Give weight to workflows with repeated searches because small waits compound across a session.

  2. Group impact by query shape

    Classify terms by length, selectivity, filters, sort order, and result volume. Mark confirmed reach from real observations separately from untested combinations. A slow rare filter may need a focused follow-up, while a common prefix or broad search can justify immediate investigation.

  3. Select one queue outcome

    Choose investigate now, schedule a scoped query or rendering improvement, or collect a missing term and dataset comparison. State the evidence that would change the decision, such as query timing by shape or a reproduction after data growth. Do not rank from the worst synthetic term alone.

What to carry forward

Produce a queue decision tied to blocked discovery, confirmed query-shape reach, and evidence quality. Include whether the delay is a regression, a scale boundary, or unresolved. The result is actionable when another engineer knows the first term and filter combination to investigate and the reason it outranks other work.

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