Impressive or is it just hype?
SevenRooms says their booking allocation process does this to “to find the best seat for every guest.”
It sounds impressive? Absolutely! Who wouldn’t want something that tirelessly reviews 10,000 combinations per second and available 24/7 for the best outcome for their restaurant!
But here’s the question nobody asks:
How does that number actually help your restaurant when a new booking request is received?
In simple terms, it doesn’t. That’s the part most booking platforms avoid talking about.
SevenRooms can show you how many bookings came in. It can show trends, reports, and totals. But when a real guest clicks “Book Now” at 7:42 pm on a busy Friday, that number does nothing to help your restaurant make a smarter decision.
At that exact moment, SevenRooms does not optimise or maximise the booking.
This is where WizButler works differently.
WizButler is not built to simply count bookings. It is designed to think at the moment a booking request is received. Instead of passively accepting reservations, WizButler actively evaluates them in real time, using data, patterns, and availability logic to decide what actually benefits the restaurant.
Problem #1: It’s Done Against a Fixed Floor Plan
SevenRooms applies its claimed combinations against a static table-based model of your venue. Tables are treated as fixed objects with predefined combinations, not as furniture within a real, three-dimensional space that changes over time. The system has no understanding of distance, spatial flexibility, or how your dining room can physically adapt in the moment. As a result, it cannot reason about moving furniture, reshaping layouts, or constructing new configurations to accept incoming booking requests. When a valid booking does not fit one of the predefined table combinations, the system has no choice but to reject it — even when the space clearly exists.
Problem #2: Previous Bookings Are Fixed
Because SevenRooms operates on a table abstraction, previously allocated bookings must remain locked in place. The system cannot reconsider or reconstruct earlier allocations when a new booking request arrives, because it lacks a complete spatial model of the venue.
As occupancy increases, the system is forced to evaluate only what remains unused, dramatically shrinking the set of possible outcomes. At that point, even small booking requests can be rejected — not because space is unavailable, but because the system is structurally unable to re-evaluate reality as a whole.
Problem #3: Not Looking at Real Solutions
True booking optimisation would require reconsidering all existing bookings and reallocating them dynamically when each new request is received.
However, the table allocation problem is mathematically classified as NP-Hard, meaning the number of possible arrangements grows exponentially as bookings increase. Attempting to search this solution space in real time is computationally infeasible. As a result, table-based systems are forced to take shortcuts — evaluating only a limited subset of possibilities within fixed constraints. These shortcuts make the system fast, but they also guarantee sub-optimal outcomes.
Problem #4: It Only looks impressive when it is not considering a new booking request
Here’s the thing that really matters: SevenRooms with its 10,000 combinations claim only looks impressive when it is not considering a new booking request. Which is after new booking requests have potentially been rejected!
That’s correct, the reality is that:
SevenRooms only checks to see if a new booking request can be added to a fixed floor plan with fixed bookings. Like Tetris, no perfect match? New booking rejected.
SevenRooms plays Tetris with your bookings after new bookings have crashed and been rejected. The game is over. Your customer was advised their was “no availability” on your widget and booked somewhere else. You’ll never see them.
It’s like using the 10,000 combinations to rearrange the deck chairs on the Titanic its not going to help you get more bookings!
You Know This Is True
You know this is true because your staff have to manually monitor your bookings and manually rearrange your bookings for optimization.
This manual intervention is not a workflow choice — it is proof that the system cannot operate autonomously. When humans are required to continuously fill information gaps, automation has already failed. Each manual adjustment further reduces the system’s ability to make effective decisions, rendering claims of large-scale optimisation increasingly meaningless.
The Real Cost: Rejections at 60-65% Capacity
This isn’t just a technical debate. It hits your revenue directly.
Restaurants using static allocation systems — including SevenRooms — typically start rejecting bookings when they’re only 60-65% full using their static table and fixed booking allocation processes.
Your dining room has space. Your widget says “no availability.”
Based on what we see, your booking system can easily be rejecting at least 10-15 bookings per week. That’s $7,000-$10,000 in lost revenue every single week.
And the worst part? You never see these customers. They tried to book, got rejected, and went somewhere else. No phone call. No walk-in. No second chance.
What Real Optimization Actually Looks Like
Real optimization doesn’t wait for rejection. It happens the moment a new booking request is received.
When you have full information about what’s being requested, that’s when you optimize — not after.
WizButler rearranges previous bookings and table layouts in real time, autonomously, the moment a booking request is received.
The system evaluates whether moving existing bookings and existing tables to create space for the new request.
No fixed floor plans. No locked bookings. No shortcuts. Completely dynamic.
The result: WizButler takes the booking instead of rejecting it.
The Difference in Practice
SevenRooms approach:
- New booking request received
- Check against fixed floor plan and fixed bookings
- No perfect fit? Rejects the booking
- 10,000 combinations do not help
- Customer books another venue
WizButler approach:
- New booking request received
- Evaluates moving tables and previous bookings to accommodate the new request
- Rearrangement occurs in real-time
- New booking accepted
- Customer booking confirmed
Same restaurant. Same tables. Different outcome.
No Manual Rearrangements. No Risk of Overbooking.
When your system handles optimization autonomously, your staff aren’t spending service time shuffling reservations. They’re focused on guests.
And because WizButler knows exactly what’s happening across your entire floor in real time, there’s no risk of overbooking. Every table is accounted for. Every booking has a place.
Just great customer experiences — and maximum revenue from every service.
The Bottom Line
10,000 combinations per second sounds impressive in marketing materials – that all.
From a revenue perspective and from an operational perspective SevenRooms can easily reject new booking requests when your restaurant is only 60% to 65% full.
That can easily be $7,000-$10,000 for a small restaurant walking out the door every week. Customers you will never see. How does that help you, your business and your customers.
If you want to stop losing bookings and actually maximize your revenue, the solution isn’t marketing hype, it is a real solution for a real-world problem.
If We’re Wrong, Please Advise
We’ve laid out the technical limitations based on how SevenRooms’ system operates. If SevenRooms believes we’ve misrepresented their technology, we welcome clarification.
In the meantime, if you’d like to see how easily your current booking system is rejecting bookings, we can show you exactly what you’re missing.
Book a Demo and see how your booking system is costing you Revenue
