Reference Management
Match a prospect to the right advocate without burning one out, with fatigue-aware cooldowns.
Included with cscron . A seat without it sees an upsell here instead of the page.
Overview
Reference Management treats your advocates as a managed pool, not a static list. When a prospect asks for a reference, it matches the request, their industry, use case, and persona, to the best available advocate, while protecting anyone who was asked recently or has hit their quarterly cap.
What it is for, on the page. Under the headline, this page says what it produces, what makes it run, and where the output lands. Every Work page reads that line from one registry, so the three answers cannot drift from the page.
Filling the pool
Add an advocate on the page: who they are, the account they speak for, the industry, use case and persona they can credibly cover, and whether they have actually agreed. Fatigue tracking needs two more numbers, the date they were last used and how many asks they have taken this quarter, against a cap that defaults to two. Everything the matcher does is computed from these fields.
Adding the same person again updates their record rather than creating a second one, so keeping the last-used date current is just re-adding them. An advocate you record as not yet willing stays in the pool and is never matched, which is the honest way to track someone you intend to ask.
The advocacy-engine skill is a separate thing and runs read-only: it reads stored signals to spot customers who have earned the ask and drafts it for review. It does not write into this pool. Who becomes a reference stays a decision you make here.
Fatigue-aware availability
Every advocate has a status: available, on cooldown (used within the last few weeks and resting), maxed (at their per-quarter reference cap), or not willing. A match only ever returns available advocates, and lists the resting ones separately so you know to wait rather than ask. Overusing a champion is how you lose one, so the cooldown is a feature, not a limitation.
How the match is scored
Fit is scored 0 to 100: an industry match weighs most, then use case, then persona. Among available advocates, ties break toward the longest-rested and least-used, so the load spreads across the pool rather than falling on one favorite. The scoring is deterministic, so the same request always returns the same ranking.
Honest when the pool is thin
A request with no available match says so plainly, and points you to widen the criteria or grow the pool, rather than promising a reference you cannot deliver or over-asking someone who is resting. The pool view summarizes how many advocates are available and which industries you can actually cover.
Recruiting before you need to
Pools run dry because recruiting happens at the moment a reference is needed, which is the worst moment: everyone willing is inside their rest period. The daily cscron run looks for the opposite case, an account that is healthy, whose realized value is accelerating, and that is not in your pool at all, and puts it in your inbox as a reference candidate. When nothing in the pool is currently askable, the same nudge says so and moves up your list.
It never nudges you to ask an at-risk customer. That is the ask that loses the customer.