Anatomy of a resource request

Every piece of demand in ProFinda is a resource request. Before you can triage them, you need to know what a good one looks like, because the quality of the request sets a ceiling on the quality of the match.

The load-bearing fields

Three fields do most of the work. Skills, expressed in the ontology, define who is relevant. Dates, a start date and expected duration, define who is available. Commitment, full-time or a percentage, defines whether someone 60 per cent booked is a candidate or not.

Everything else, grade range, location, client, engagement code, narrows or contextualises. Useful, but a request with strong skills and dates and nothing else will still produce a workable shortlist. A request with a beautiful description and no start date will not, because availability cannot be evaluated against “soon”.

What the free text is for

The description field is for humans: the engagement context, the client situation, why this role is interesting. Matching does not read it. This trips people up in both directions. Partners write “must have insurance audit background” in the description and wonder why non-insurance people appear; the skill was never added as a structured field. Conversely, some requesters cram twelve skills into the structured list when only three are essential, and the match dilutes across all twelve.

Reading a request in thirty seconds

When a request lands in your queue, scan in this order: dates first (is it real and near?), skills second (are they specific enough to match on?), commitment third (is the ask feasible?). A request that fails any of these goes back for clarification before it goes anywhere near a shortlist. The lesson on clarifying vague requests covers how to do that without friction.

Where requests come from

Requests arrive from three routes: raised directly in ProFinda, created from your CRM when an opportunity passes a probability threshold, or entered by your team on a partner’s behalf after a corridor conversation. The third route is the one to watch; requests transcribed from memory are where the missing start dates come from.