Until now every booking, refund and order carried its own separate copy of the customer’s details, and nothing joined them up. A person you had dealt with twenty times was a stranger on the twenty‑first, and the AI had to ask for an email address on every single call — the least reliable thing that happens on a phone line. There is now one record per customer, matched on their email address or phone number, and the AI recognises whoever is calling from the number they are ringing from and uses what you already hold. Records are created for you as calls come in, and everything that customer has ever booked, ordered or asked for a refund on sits on one page. You can add your own fields too — a flat number, a boiler model, a policy reference — and the AI can read those on a call.
What changed
The customer database joins operational records around one customer profile. Names and contact details can be kept alongside bookings, orders, refund requests and the context your team needs when the person contacts you again.
Custom fields make the profile relevant to the business: a property detail, equipment model or policy reference can sit with the customer rather than being repeatedly typed into unrelated notes.
Before and after
Before this update: Bookings, refunds and orders carried separate copies of contact details. A returning caller’s history was not joined around a single shared customer record.
With this update: Customers are matched by email or phone number, records can be created as calls arrive, and the receptionist can use the details already held when it recognises a caller.
What it means for you
Check existing customer details and choose custom fields that help your team answer real enquiries. Open a profile to review the connected records instead of searching each operational page separately.
Recognition helps find the right record; it does not remove the identity checks needed before discussing protected details. Keep contact information accurate so shared records remain useful.