Enquiry replies
The enquiry at 11 pm gets the same answer as the one at 11 am.
Most enquiries are the same twenty questions: price, size, availability, when, where. The answers are in a price list or a catalogue. They just arrive at night, at weekends, in another language, or while your team is with someone else.
WhatsApp · sales enquiry
visit booked, sales told
What this produces, shown with sample data.
The short version
It answers those on WhatsApp, on your site or on the phone, from your real data, in the customer's language. It books the visit or the call, and hands anything unusual to a person with the conversation already summarised.
It does not negotiate, promise, or quote what isn't in the price list. Those stay with your team.
Where it runs
Where this kind of tool runs today, and where we'd build it next.
| Industry | Job | Status |
|---|---|---|
| Retail | Thai-language shopping concierge for a supermarket chain | Live |
| Real estate | WhatsApp enquiry reply for off-plan launches | In build |
| Insurance | First notice of loss by phone or WhatsApp | Playbook ready |
| Property | After-hours maintenance requests | Playbook ready |
Start to finish
- 1
Write down correct
We collect a few hundred real enquiries from your inbox and chat history. Each one gets the answer a person on your team would sign. That set becomes the test it is graded on.
- 2
Build inside your tools
It works where the work already happens: WhatsApp, email, your POS, CRM or practice software. Nobody learns a new tool.
- 3
Run it with a person approving
For the first weeks a person approves what it sends. We loosen that only where the scorecard says it has earned it.
- 4
Grade it every month
The test re-runs monthly and whenever the model underneath changes. You get the score and the list of what it got wrong.
Measured in minutes, not messages
We count minutes to first reply and visits or calls booked, not how many messages it sent. A chatty tool is not a useful one.
If nobody can write down what correct looks like, we don't build it yet.
Tell us the job your team hates. We'll tell you if AI can do it.
A 30-minute call with the people who would build it. No technical knowledge needed: just describe the job, how often it happens, and how you check it's done right.