Quote requests, handled without losing the human touch.
Who it's for: an operations manager at a small trade or service business who answers quote requests by email. Pick a sample request and walk it through the four steps: the request arrives, a draft is prepared, a person checks it, and the record is updated.
⚠ Internal demo with made-up sample data. The business, customers and replies are fictional. Nothing is sent anywhere.
1Request arrives
Inbox
2Draft prepared
AI · draft only
3Person checks
Human review · required
4Record updated
Lead tracker
Waiting for the person to approve.
Received
Customer
Job
Area
Status
Next action
Where a person must stay in charge
Prices and dates
The draft never commits to a price, a start date or a deadline. Those come from the person who knows the diary and the costs.
Anything unusual
Safety concerns, complaints, legal or insurance questions, and anything the draft isn't sure about are flagged for the person, not answered.
Sending
Nothing goes to the customer until someone reads it and presses approve. The draft is a starting point, not the reply.
Systems this pattern works with
Where requests arrive
Gmail or Outlook
Website contact form
WhatsApp Business
Where the draft is made
Claude, ChatGPT or Copilot with your saved instructions
Or an automation tool (Zapier, Make, n8n) that passes the email to AI
Where the record lives
A Google Sheet
Baserow, Airtable or a simple CRM such as HubSpot Free
What this demo is and isn't. It shows the shape of the workflow with labelled sample data so you can see where the human checks happen. The drafts here are written examples, not live AI output, and no time or money savings are claimed. Set it up in your own business with the step-by-step playbooks below.
Build it in your business
The Systems Vault walks you through it: "Instant reply to every enquiry" and "Website enquiries straight into a lead tracker", plus 15 more playbooks. Try a free sample first.