Enquiry to CRM record and booked appointment

Demonstration, not a client case study Professional services Sales Standard · 2 credits

Before

Enquiries arrive through the website form and a shared sales inbox. Someone copies the details into HubSpot by hand, often hours later and sometimes twice. Replies depend on who is at their desk, and the calendar link goes out in a separate email that is easy to forget.

Trigger

New website form submission or inbound email in the sales inbox

Connected systems: HubSpot, Outlook, Calendly, Slack

Automated steps

Classify the enquiry: request type, urgency, language (AI step, confidence threshold)
Enrich company data from the email domain and check for an existing HubSpot contact or company
Create or update the HubSpot record, assign the owner by region and service line, no duplicates
Draft a reply in the owner's tone with the matching Calendly link (AI step)
Post the draft and the record summary to the owner's Slack channel for approval
On approval, send the reply from the owner's Outlook mailbox and log it in HubSpot
When the Calendly booking is made, update the deal stage and send the owner a short summary

Human checkpoint

The account owner sees the classified enquiry, the enriched record and the draft reply in Slack. They approve, edit or reject. Nothing reaches the prospect without this step. Low-confidence classifications and unknown domains go to the same queue flagged for manual handling.

Output

A clean CRM record with owner and source, a reply sent within the working hour, a booked meeting on the right calendar and a logged summary for the owner.

What is measured

First-response time
Hours per week spent on manual CRM entry
Duplicate contact rate
Enquiry-to-meeting conversion

Risks and limitations

Misclassification of unusual enquiries, mitigated by the confidence threshold and approval queue
Enrichment data can be stale or wrong for small companies
Personal data is processed; retention and minimisation rules must be set in HubSpot
Calendly and HubSpot API changes require monitoring

Demonstration using sample data — not a client case study

Walkthrough

A prospect fills in the contact form at 09:12 asking about “process mapping for our finance team, fairly urgent”. The workflow reads the submission, classifies it as a services enquiry with high urgency in English, and looks up the email domain. It finds no existing HubSpot contact but does find the company, so it creates the contact under the existing company record and assigns the owner responsible for that region.

The AI step then drafts a short reply in the owner’s usual tone, referencing the finance-team context and including the owner’s Calendly link. At 09:14 the owner receives a Slack message with the classification, the record summary and the draft. They change one sentence and approve. The reply is sent from their own Outlook mailbox and logged against the contact in HubSpot.

When the prospect books a slot later that morning, Calendly’s webhook moves the deal to “meeting booked” and the owner gets a two-line summary with the enquiry text and the booking time.

Where the workflow is uncertain, it does less, not more. An enquiry it cannot classify with confidence, or a domain it cannot match, lands in the same Slack queue flagged for manual handling. The person still sees every case; the difference is that the routine ones arrive ready to approve.

Next step

Start with one workflow worth fixing

Answer nine short questions and get an honest read on whether your process is a good automation candidate — and which package fits. No sales call required.

Run the free Opportunity Scan

Book a 20-minute fit call