The problem
Apex Fintech grew through two acquisitions, and each brought its own CRM export. Combined, the lists held about 3,100 contacts, several thousand of which sat in jurisdictions Apex wasn't licensed to sell into yet. AEs were qualifying prospects by hand against a compliance checklist before a conversation ever started, and a mistake there was expensive to unwind.
The company-size and geography factors in a default scoring model weren't strict enough for Apex's situation - a well-funded prospect in the wrong jurisdiction still isn't a prospect, no matter how good the rest of the fit looks.
What they configured
The two CRM exports were imported as separate batches; duplicate detection on email and LinkedIn URL removed roughly 400 contacts that existed in both. Apex raised the geography weight and lowered outsourcing probability, since regulatory fit mattered more to their pipeline than budget signals did, and moved the qualification threshold to 78% - above the platform default - to keep the queue narrower than usual.
Six outreach owners were set up on criteria-based distribution, one per licensed region, so a prospect in a jurisdiction Apex couldn't service never reached an owner's queue in the first place rather than being screened out manually after the fact.
What happened
Raising the threshold to 78% produced a smaller queue, but a materially cleaner one - the share of conversations that turned out to be a jurisdictional dead end fell by roughly three-quarters. Owners spent the time that used to go on compliance triage on the outreach sequence itself.
Criteria-based routing put over half of every day's queue in front of the owner who actually covered that prospect's region, with the daily limit capped per owner so no single territory could crowd out the others. The AI-drafted opener referenced Apex's regional case studies automatically, since the routing rule had already matched the right owner to the right regional content before a person saw the prospect.