Part 1 The problemWhy teams need this
01 · THE PROBLEM
Triage decides how fast a claim moves, and most teams still do it by reading claims in arrival order. Scoring severity at intake means the critical claim filed at 4:55pm doesn't wait behind forty minor ones.
Part 2 How it worksWhat it does, step by step
02 · WHAT IT DOES
Claims come in by email or API. The workflow extracts the details, checks the policy, assesses severity, looks for fraud indicators and routes each claim to the right adjuster team, with every decision written to an audit log for your regulators.
03 · HOW IT RUNS
Step by step, as built.
Normalise intake
Email and API claims are put into one shape and merged.
Extract and verify
Claude extracts the claim details and the policy is checked in Salesforce. Lapsed policies get a reply, not a queue slot.
Assess severity
Severity is scored against a fixed schema, so every claim is judged on the same criteria.
Check fraud indicators
A rules step flags suspicious patterns and alerts the investigations team.
Route and audit
A Switch assigns critical, standard and fast-track claims to the right team; every decision is written to the audit log and the claimant is acknowledged.
04 · WHERE A PERSON STAYS IN
The machine drafts. A person decides.
Adjusters make every coverage decision. Suspected fraud goes to your investigations team, never to an automatic rejection.
05 · TOOLS AND APPS
Built around the systems already in the process.
06 · WHEN SOMETHING BREAKS
Failure is designed in.
- 10nodes retry automatically when an external API fails.
- 3decision points (IF or Switch) check the data before it moves on.
- ✓Anything unhandled triggers our central error workflow, so a crash gets reported instead of failing silently.
Standard on every build
- Schema validation before downstream writes
- Retry and error routes for external API failures
- Duplicate-safe processing and idempotent updates
- Human approval where the action carries business risk
- Execution logging for support and audit review
Part 3 The impactWhat it's worth, and how we'd build yours
07 · PROJECTED IMPACT
What it should change in the business.
Projections for a typical deployment. The calculation below shows the math, and you can put in your own numbers.
08 · ROI CALCULATION
How it pays back in your business.
The starting numbers are a hypothetical deployment sized to the projections above. Change any of them to your own volumes and costs, and the math updates underneath.
- Full-time equivalent freed
- 0.8 people
- Gross value
- $57,600
- Running cost
- −$960
- Return per $1 of running cost
- $60
The math: 600 claims triaged × 20 min × 12 months × 60% ÷ 60 = 1,440 hours a year × $40/hour = $57,600. Net value = gross value − $960 running cost a year.
09 · HOW WE'D BUILD YOURS
How we'd build yours.
- Discover: map the current process, systems, volumes, owners and exceptions.
- Design: define the canonical data model, approvals, retries and system boundaries.
- Build: implement credentials, nodes, validation and observable error routes.
- Prove: run controlled data through success, duplicate and failure scenarios.
- Operate: publish runbooks, ownership and measurable service levels.
