In this article
Every booking that arrives comes in through a door — and most restaurants, without ever counting, only have one door that's actually doing the work.
A restaurant can be booked six different ways today: the phone, its own website, the reserve button on Google, the reserve button on Meta, a marketplace app such as TheFork, Quandoo or Zenchef, and a plain WhatsApp message from a regular. Opening each new door feels like a win — more visibility, more ways for a guest to say yes. What almost nobody adds up is what happens the moment one of those doors ends up carrying well over half of all bookings. That stops being a visibility question. It becomes a risk you can measure, with a tool that has nothing to do with hospitality.
Six Doors, One Booking Book
The same question applies to each channel: what does it cost you, and how much control do you actually keep over it? That second part is the one almost always skipped.
1. Phone
The oldest channel, and at many venues still the biggest silent one. It costs no commission, but it costs staff time: a phone that rings three times during peak service and goes unanswered is a booking that just went to the place next door. You own the conversation entirely — but none of it lands in your system unless someone types it in.
2. Your Own Website
The one channel you genuinely own end to end: the conversation, the guest's details, the repeat booking with nobody standing in between. No commission per cover, no algorithm deciding whether you're visible today. The price is that nobody sends guests here except you — this channel does not grow on its own.
3. Google (Reserve with Google)
Appears right next to your search result, at the exact moment someone has already decided to eat out and is only looking for a table. Free to connect, but Google decides which booking widget it shows next to your name, and can change that choice without asking.
4. Meta (Facebook & Instagram)
A reserve button on a profile you're already maintaining for photos and stories. Small volume for most venues, but the one channel where a booking comes directly out of social content — without that one post about the new menu, that table would never have existed.
5. A Marketplace App
TheFork, Quandoo, Zenchef and similar platforms hand you the single biggest visibility boost available — and build the single biggest dependency at the same time. You pay per booking or per month, sit on the same page next to three competitors, and the platform sets the rules: the commission, your ranking inside its search results, and whether your account is still active next month.
6. WhatsApp
Informal, direct, and for many younger guests already the first channel that comes to mind: just send a message. No commission, no queue — but also no automatic link to your table plan, which is exactly why a double-booking is easiest to create right here.
A Familiar Mix (Illustrative)
Not measured data — an example split the way many independent venues report their own. Work out your real numbers further down in the score.
In this split, more than half of all bookings sit on one platform the venue doesn't own.
So six doors is no guarantee of spread — it's just as possible for one of them to have quietly swallowed all the others. The question isn't how many doors you have. It's how heavily the book leans on one of them, and there's an exact number for that.
The Concentration Number: How Regulators Measure Dependency
A tool for exactly this question already exists, and it doesn't come from hospitality — it comes from competition economics. When regulators such as the US FTC or the European Commission review a merger, they look at the Herfindahl-Hirschman Index (HHI): every player's market share, squared, added together. A market with ten equal players at 10% each scores 10 × 10² = 1,000. One player with 100% scores 10,000. The higher the number, the more concentrated the market — and the more exposed it is to that one player disappearing.
The same sum works on a booking mix. Take each of the six channels above as a share of the total, square each percentage, add them up. The US merger guidelines draw the line at 1,500 (unconcentrated) and 2,500 (highly concentrated) — but those thresholds assume a market with dozens of players. With only six realistic doors, even a perfectly even split (16.7% each) already scores 1,667: above the first line before anything has even gone wrong. For a booking mix, the bands shift accordingly: under 2,500 is spread, 2,500 to 4,000 is concentrated, above 4,000 is heavily dependent — the scale the calculator further down uses.
Take the example mix above: 12-15-10-5-52-6. Square each percentage and add: 144 + 225 + 100 + 25 + 2,704 + 36 = 3,234. Concentrated, well past the first line — purely because one channel carries more than half the total.
What Happens When One Disappears
An HHI score stays abstract right up until the platform it's built on does something outside your control: raises its commission, changes its ranking algorithm so the venue drops lower, freezes an account over a billing dispute, or — this has already happened elsewhere in hospitality with comparable consumer platforms — shuts the product down outright. What that actually costs can be worked out by running the identical shock through two venues with different spreads: each one's biggest channel goes dark for a week.
Same Shock, Two Outcomes
The biggest channel goes dark for a week — an outage, a dispute, a delisting. What's left of that week's bookings?
Same shock, a different weekend: a 55-percentage-point gap, entirely down to the spread that already existed before the outage.
Nobody deliberately plans a week around one channel — it happens gradually, because the channel that brings in bookings the easiest is also the one you have the least leverage against. Which is exactly why it needs to be measured rather than felt.
Work Out Your Own Concentration Score
Enter roughly how many bookings arrive per week through each channel — a rough estimate is fine. The tool works out the total, each channel's share, your own HHI, and how many bookings are on the line if your biggest channel disappears tomorrow.
The Concentration Score
Bookings per week, per channel
Total bookings/week
100
Your concentration number (HHI)
2200
SpreadNo single channel carries the majority. If one goes down, the rest holds up the book.
30 bookings a week on the line if your biggest channel disappears tomorrow
A score above 4,000 isn't a reason to drop a channel outright — it's the reason to start building a second real door before the first one decides to do that for you. See /en/blog/reservations/increase-direct-bookings-restaurant.html for concrete tactics to grow the one channel you actually own.
What You Can Already Do This Week
This doesn't call for rebuilding your whole booking strategy — just four habits that keep the spread healthy:
- Keep at least two active channels running side by side, and check that both write to the same table plan. A second channel that doesn't sync solves nothing — it just adds a double-booking risk.
- Score above 4,000? Start building the channel you actually own this quarter: your own site. One line on the receipt or the menu ("book direct via [name], no commission for us") shifts bookings without asking the guest to do anything extra.
- Never let the phone roll to voicemail during peak service. The cheapest channel you have loses bookings at exactly the moment you need them most.
- Recalculate the score every quarter, not once. A marketplace app carrying 20% of your bookings this year can carry 45% next year without you ever consciously asking for "more" — it happens on its own the moment it becomes your easiest channel.
Conclusion: More Visibility Isn't the Same as Less Risk
Six doors feels like safety, but safety was never in the number of doors — it's in what happens the moment one of them locks. A marketplace app raising its commission or changing its algorithm tomorrow stops being a hypothesis the moment you know that one app already carries 60% of the booking book. It's a sum you can do today, with numbers you already have in your head.
The sixth door isn't even the last one: /en/blog/reservations/chatgpt-books-restaurant-tables.html covers how AI agents are starting to book tables on a guest's behalf — a seventh channel taking shape, carrying the exact same question. And anyone still weighing booking software rather than booking channels will find that comparison in /en/blog/reservations/online-reservation-system-pros-cons.html. Both are worth reading, but neither answers this one: how much of today's revenue already sits in the hands of a platform you don't own?