Part 1 The problemWhy teams need this
01 · THE PROBLEM
The real customer history lives in inboxes, not the CRM. So the CRM is always out of date, and the person picking up a deal has to read forty emails to catch up. An agent that files as mail arrives turns the inbox into a live record, and makes sure a hot reply never sits unseen.
Part 2 How it worksWhat it does, step by step
02 · WHAT IT DOES
Every client email is matched to the right person and the right open deal by an AI agent, saved to the CRM, summarised, and classified by intent. Interested leads alert the deal owner, confirmations advance the stage, and clear declines get a polite close. When the agent isn't sure who an email belongs to, it asks a human instead of guessing.
03 · HOW IT RUNS
Step by step, as built.
Skip the noise
System and no-reply senders are filtered out first.
Match the sender
The contact is found by email; unknown senders raise an alert instead of a wrong match.
Attribute to the right deal
Claude reads the email and picks the right open deal against a strict schema.
Summarise and save
The email is saved, key facts are extracted and the deal summary is updated.
Act on intent
A classifier routes interested, confirmed, not-interested and other replies to the right action.
04 · WHERE A PERSON STAYS IN
The machine drafts. A person decides.
The agent only files when it's confident. Unknown senders and ambiguous matches go to Slack for a person to decide.
05 · TOOLS AND APPS
Built around the systems already in the process.
06 · WHEN SOMETHING BREAKS
Failure is designed in.
- 11nodes retry automatically when an external API fails.
- 2decision 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
- 0.4 people
- Gross value
- $30,600
- Running cost
- −$1,080
- Return per $1 of running cost
- $28.3
The math: 1,500 client emails × 3 min × 12 months × 85% ÷ 60 = 765 hours a year × $40/hour = $30,600. Net value = gross value − $1,080 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.
