Pilot scope
- Input
- 500–5,000 historical claims from your CMS, with documents and decisions.
- Scope
- One workflow, pre-auth, discharge authorization, or reimbursement. Scoped in week 1.
- Deliverables
- Reviewer dashboard · document extraction · claim summary · missing-doc detection · query drafting · TAT measurement.
- What we do not do
- We do not auto-approve claims. We do not access data outside the agreed scope. We work under NDA from day one.
Target outcomes
- 30 to 50%Lower review time per packet
- 20 to 40%Fewer avoidable queries
- 1.5 to 2xReviewer productivity uplift
- 100%Claims with evidence trail
Timeline
Six weeks, three checkpoints.
-
WEEK 01 - 02
Scoping and setup
Workflow selection, success metrics, data access, and review of historical claims. We map your current process step by step.
-
WEEK 03 - 04
Configuration and dry run
We tune document extraction, summary templates, and query drafts on your historical claims. Your reviewers validate the output.
-
WEEK 05 - 06
Live workflow and readout
Reviewers work on live claims through the Slate Labs dashboard. We measure TAT, queries, and productivity, then present a written readout.
What we ask from you
Lightweight, but real.
An executive sponsor
Usually the COO, Head of Claims, or CMO. Someone who can clear data access blockers, align the review team, and treat the pilot as a real operational priority, not a side experiment.
2 to 5 reviewers
A small group from your medical or claims team who will test the dashboard, give feedback on the agent outputs, and run live cases in weeks 5 and 6. Their feedback shapes the final readout.
Sample claim files
500 to 5,000 historical claims with the original documents and final decisions. We measure our outputs against your real outcomes, not synthetic benchmarks.