Part 1 The problemWhy teams need this
01 · THE PROBLEM
Nobody writes good call notes after the fifth call of the day, so the most valuable information a sales team collects, what the customer actually said, mostly disappears. Capturing it automatically makes every follow-up sharper and every handover painless.
Part 2 How it worksWhat it does, step by step
02 · WHAT IT DOES
When a call ends, the workflow fetches the transcript, matches it to the booking it belongs to, and works out whether it was a real call or a no-show. Real calls go to an AI that writes the summary, the action items and the customer's pain points, which are saved to the CRM and posted to Slack. Calls it can't match raise an alert.
03 · HOW IT RUNS
Step by step, as built.
Verify and fetch
The webhook signature is checked and the transcript is pulled from the meeting platform.
Match to a booking
The call is tied to its booking; a missing match raises an alert rather than a lost call.
Real call or no-show?
Duration and participants decide it. No-shows are marked and the host is told.
Read the call
Claude writes a summary and action items against a fixed schema, then pulls out the customer's pain points.
Write it down
The call record is saved, the booking status updated and a summary posted to Slack.
04 · TOOLS AND APPS
Built around the systems already in the process.
05 · WHEN SOMETHING BREAKS
Failure is designed in.
- 11nodes 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
06 · 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.
07 · 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
- 0.3 people
- Gross value
- $21,600
- Running cost
- −$840
- Return per $1 of running cost
- $25.7
The math: 200 sales calls × 15 min × 12 months × 90% ÷ 60 = 540 hours a year × $40/hour = $21,600. Net value = gross value − $840 running cost a year.
08 · 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.
