Reply, follow up and stop over-messaging.
AI Engagement sets up a chat assistant on your website and messaging channels that answers from approved content, drafts replies for staff and keeps follow-ups on a schedule. Anything it cannot answer, and any upset customer, goes to a named person who reads the whole conversation before replying, so nobody repeats their story. It supports staff, never replaces them.
Questions arrive when nobody can answer. A salon, gym or restaurant hears about opening hours, package prices and Saturday tables late in the evening, and many of those people never ask a second time.
An assistant can answer the routine questions from content you approved, and take a booking request or a callback number for the rest. Staff then see which channel each customer prefers, when they reply and what they ask about, so the next message fits the person instead of the whole list. Every reply stays on record.
The problems this is meant to ease.
What AI engagement does.
What this looks like in a working week.
The people who open this every day.
From first conversation to daily use.
Discovery
List every channel first. Owner, reception and marketing lead sit down with us, and together we list the channels where customers write, the questions they ask most and who answers them.
Data readiness
Write the answers down first. The common questions get approved answers, customer contacts are gathered into one list with the date and source of each permission, and past chats are read for gaps.
Trial on website chats
Start with the website chat. It answers on the website chat alone with a staff member watching every conversation, and nothing is sent to the customer list during this stage.
Front desk review of live chats
Staff mark the wrong answers. Support agents and reception staff read real conversations, mark answers that were wrong or cold in tone, and each mark becomes an edit to the library.
Going live
Add channels one by one. Messaging apps and email follow the website chat, win-back messages start with one small group, and the one-switch pause stays within reach of the manager at all times.
Reviewing afterwards
Read the weekly lists. The team reads unanswered questions, reply times and opt-outs after a full cycle, adds missing answers, retires messages that annoyed people and compares each measure with its baseline.
How you would know it is working.
Record where you stand before it goes live, so the change is measured against your own figures.
- Engagement rate counts opens, clicks and replies against messages delivered, and the baseline is that count taken from your email and messaging tools for the campaigns before go-live.
- Response rate counts customer messages answered by a person or the assistant, and the baseline is the answered count from your inbox and chat records before go-live.
- Lead nurturing conversion counts enquiries that became bookings or sales after follow-up messages. The same count, taken from your CRM for the three months before go-live, is the baseline.
- Handover rate counts conversations passed to a person, and the baseline is the number of chats your staff handled by hand in the weeks before go-live.
- Opt-outs and complaints are logged after every send, and the baseline is the count from the last campaigns you sent to the whole list before frequency limits began.
Where a person stays in charge.
A person owns the answer library
Someone must own the updates. Replies come only from the library your team approved, so a wrong answer is a wrong entry that a named person fixes, and early chats are read.
Upset customers reach a person
Sensitive topics go to staff. Refund requests, complaints, health or allergy questions and anything outside the library go to a named staff member who receives the whole conversation, and the assistant stops replying.
Consent and opt-outs
A reply of stop is final. Each customer’s permission is saved with its date and source, and an opt-out removes that person from every later send on every channel.
Messaging rules differ by country
Rules are set by country. Duties for direct messaging, template approval on chat apps and keeping chat records vary by country and platform, so your own lawyer should confirm each. It prepares records, not legal advice.
Limits of the assistant
Staff keep every decision. It cannot judge tone, take payment or promise a refund, and staff can pause it on every channel with one switch for a holiday or an away day.
It is one part of a system, not an island.
A module earns its place by what it passes to the next one. These are the connections we set up most often.
- Website chat
- WhatsApp Business messaging
- Booking or appointment software
- Point of sale or billing
- Customer records in your CRM
- Email and SMS tools
How we would put it in.
Questions owners ask about AI engagement.
Will it say wrong things?
It can still be wrong. It answers only from content you approve and hands over anything else, which limits the risk without removing it, so someone must own updates and review early chats.
Do we need WhatsApp for this?
A website chat alone is enough. WhatsApp can be added later through the official business interface, which has its own message approval steps, and rules vary by country and by the platform’s own terms.
Which data does it use?
Start with approved answers. Your customer contacts with the date of each permission, a booking or billing system if you have one, and your usual questions are enough to begin.
Is a loyalty scheme worth it?
Only where visits repeat. As at a salon, a cafe or a gym, points matter when visits repeat, while a business that sells once in years gains little.
What about privacy and permission?
Rules depend on the country. Consent and direct messaging duties differ by market, so your own lawyer signs off how permissions are captured, and the software only prepares records. It gives no legal advice.
Can staff pause it?
Yes, with one switch. It pauses on every channel and a short notice tells visitors that a person will reply, which suits holidays or a day when stock is uncertain.
When is this not worth it?
When few people write in. A shop that gets only a handful of messages does better with a person and a phone, since the library only makes sense once volume rises.
What effort does setup take?
Mostly writing and tidying. Someone writes the approved answers and the reward rules, the contact list is cleaned up with permission dates, and then channels are connected one at a time.
Businesses that ask for this.
Modules that work with it.
Tell us how you handle AI engagement today.
A spreadsheet, a register or another system: say which, and we will tell you plainly what is worth changing.
- No obligation
- A reply within one business day
- Your details stay private
Start with your requirements
Seven short steps that tell us what you need. An engineer reads them before getting in touch.
Share your requirements