n8n
85% of guest messages, handled alone
A family-owned 45-room boutique hotel with six staff was drowning in guest messages: about 1,200 a month across Booking.com, Airbnb, Expedia, email, and WhatsApp, a 4.2-hour average reply time, 18 admin hours a week, and 3 to 4 double-bookings a month. Builders connected all five channels into one inbox. An AI system built on n8n sorts each message (check-in, breakfast, complaint, upsell, FAQ), syncs availability every 15 minutes, sends a pre-arrival WhatsApp upsell, and requests a review two hours after checkout. The build took 5.5 weeks and costs about £180 a month, per the builders.
How it wentIn 90 days the system handled 85% of messages on its own and staff saw only the flagged 15%, per the builders. Reply time fell to under 90 seconds, admin fell to under 3 hours a week, double-bookings hit zero, and monthly revenue rose from £54,000 to £69,120. The property is anonymised and the figures are the builders' own report.
Try it · based on the source's workflow
Run guest messaging for [hotel name] from one inbox covering [channels]. Sort each message as check-in, question, complaint, upsell, or FAQ. Answer FAQs and check-in questions only from our facts: [paste]. Any complaint, refund, or special request goes to [staff contact] at once with the full thread. Two hours after checkout, send a review request in our tone: [example]. Send me a daily summary: messages handled alone, messages flagged, and anything a guest is still waiting on.
A starting template based on the workflow the source describes. Review it before you use it; you are responsible for what it does.