Common search problems

Work through these in order; they are sorted by how often they turn out to be the cause.

1. The skill was free text, not an ontology skill

If a skill was typed into a request or profile without picking from the suggestion list, it is unmatchable text. Fix: edit the field, retype the skill, and select from the suggestions. If nothing suitable is suggested, the skill may genuinely be missing - raise an ontology query via your platform team rather than inventing a workaround.

2. An expected person is missing from a match

Almost always a hard-constraint filter, not a skills issue. Check in order: their availability in the request window, the grade range on the request, location or mobility settings, and conflict rules. Filtered people are excluded silently; the match breakdown of people who did appear tells you which constraints are active.

3. Results look right but rank oddly

Read the score breakdown before assuming a fault - availability weighting commonly explains it. See the match score and the lesson When matches look wrong for the full diagnostic routine.

4. Search returns too much

Broad skills (“project management”) match half the firm. Use two or three specific skills together, then apply grade and availability filters. If you are exploring capability rather than staffing a role, use the search page, not a request - requests are for real demand.

5. New joiner or recent change not appearing

Profile and availability data arrive via integration feeds and can lag by up to 24 hours. Check the person’s profile page directly; if the change is visible there but not in search results, the search index is behind - it rebuilds hourly, so wait, then raise a ticket if it persists past a day.

If none of these fits, raise a ticket including the request URL, the person affected, and a screenshot of the breakdown panel.