Restaurants are flying blind. A four-star Google review tells an owner their customer was vaguely happy. It doesn't tell them the carbonara has gotten salty since the new line cook started, that the tiramisu is the best dish on the menu but nobody's ordering it, or that the host was distracted on Friday night. The signal that would actually change a shift, a recipe, or a hire never reaches the kitchen.
Bibi is the opposite of a review aggregator. Instead of broadcasting opinions to strangers on the internet, it captures granular, item-level feedback the moment after the meal — and sends it only to the owner.
How it works.
A QR code on the receipt opens a mobile page styled like the menu itself. The guest taps the dishes they ordered, gives each one a star and an optional comment, rates the evening overall, and drops their phone number. In return, they get a one-time discount code by text for their next visit. No app, no account, no friction.
What the owner sees.
A dashboard with per-item ratings, comment streams, and trend lines — plus an on-demand AI weekly briefing that reads every comment, finds the patterns, and writes them up in plain English: what to fix this week, what's working, what to push. Specific enough to act on Monday morning, private enough that none of it ends up on Yelp.
What makes it different.
Three things: (1) the data is item-level, not experience-level, so it points at a dish or a shift instead of a feeling; (2) it's private intelligence for the operator, not a public review feed; (3) the AI summary translates a hundred messy comments into a punch list. Specific. Private. Actionable. That's the whole pitch.
Built with Next.js 16 (App Router, Turbopack), Tailwind v4, Supabase, Twilio, and Claude Opus 4.7 for the weekly summary.
Built With
- next.js
- supabase
- tailwind
- twilio
- vercel
Log in or sign up for Devpost to join the conversation.