All case studies
Hospitality
Restaurant reservations & feedback loop
Confirm, remind, seat, follow up. One workflow, one WhatsApp thread per diner.
Client: AI Commerce OS
n8nGemini 3 FlashWhatsApp Cloud APIGoogle CalendarSendGridPostgres

The challenge
A restaurant group with four locations was juggling bookings from Google, Instagram DMs and the phone, with no single view of any of it. Reviews after a visit were hit-or-miss, and marketing had no clean way to nudge diners who'd stopped coming back.
Solution design
Every booking lands in one shared calendar. Two hours before the reservation, the diner gets a WhatsApp confirmation. About 90 minutes after they leave, they get a quick prompt: if it's a 4 or 5 we hand them the Google review link, anything lower goes straight to the manager. A separate segment picks up diners who haven't been in for a while and drops them into a small promo campaign.
Architecture
flowchart LR
A[Booking Sources] --> B[Unified Calendar]
B --> C[T-2h Confirm]
C --> D[Post-visit +90m Review Prompt]
D --> E{Rating}
E -->|4-5| F[Google Review Link]
E -->|1-3| G[Manager Follow-up]
B --> H[Lapsed Diner Segment]
H --> I[Promo Campaign]Rendered as a Mermaid flowchart in the deep-dive report.
Workflow screenshots
Reservation thread
Reservation thread
One WhatsApp thread from booking to review, per diner.
Lapsed-diner promo run
Lapsed-diner promo run
Segment auto-refreshes daily; 30-day cooldown per diner.