This is the shortest honest course in the library, because for most restaurants the answer is no. Somebody will offer to build you an app, the demonstration will be beautiful, and the arithmetic underneath it almost never works for a single fifty-seat restaurant. This first topic is that arithmetic, done in public, so that you can either close the course with your money still in your pocket or carry on knowing exactly what you are buying.
📱 Three conditions, and all three have to be true
An app earns its place only when all of the following hold at once. Two out of three is a no.
- People come often enough. Installing an app is a cost the customer pays — finding it, downloading it, making an account — and they only pay it for places they use repeatedly. A coffee shop somebody visits four times a week clears this easily; a restaurant somebody visits every six weeks does not, no matter how much they like you.
- There is something the browser genuinely cannot do. This list is much shorter than it was: a mobile site can take orders, take bookings, take payment, show a menu and hold a loyalty card. What is left is essentially notifications you can send for free and a home-screen icon — real advantages, small ones.
- Somebody will look after it forever. An app is not a thing you buy, it is a thing you maintain: the menu changes, the phone operating systems update twice a year, the store rules change, the payment library expires. An app nobody maintains stops working within about eighteen months, and by then it is doing you damage rather than nothing.
Notice that none of those three is about design or features. The catalogue version of this subject spends its chapters on which features to include; the decision was already made or lost before that conversation starts.
📉 The install arithmetic, with your own numbers
Run the funnel for the restaurant this library has been using. It serves two thousand covers a month and has, generously, eight hundred people who could be called regulars.
Installs. A restaurant that promotes an app well — on the check, at the table, at the door — typically converts a low single-digit percentage of its regulars. Call it 5%: forty installs. That is not pessimism, it is what a small independent gets without a permanent marketing budget.
Still there after a month. Most apps lose more than half their users within thirty days. At 40% retention that is sixteen people still opening it.
What sixteen people are worth. If each of them orders once a month at a $25 check, the app is moving $400 of sales — 0.8% of the restaurant's month. Against a white-label app at $300 a month, that is $18.75 of app cost per order, and the fee eats three-quarters of the sales it produces.
And the fee is a fixed cost, which the budgeting course already taught you how to read: $300 a month divided by the 51% that survives to cover fixed costs means the app has to generate about $600 of extra sales every month just to be neutral. Sixteen people ordering once will not do it.
🪞 The mirror test
Before any of the numbers, there is a two-second version of this decision that is remarkably accurate: how many restaurant apps are on your own phone, and when did you last open one?
Almost everybody's answer is one or two — usually a big chain with a genuinely good points programme — and last time was months ago. Your customers are not different from you. They will install an app for a place they use weekly and a brand that gives them something specific; they will not install one for a very good restaurant they visit six times a year, and being annoyed about that does not change it.
The chains are the exception that proves the rule, and it is worth seeing why: they have the frequency, they have a real reward programme, they have people whose whole job is that app, and they have millions of customers to spread the cost over. If your app is trying to be a small version of theirs, you are competing on the one axis where scale wins completely.
🧭 What this course covers, and what it will not repeat
Four of the five chapters the original outline proposed for this subject belong to other courses, and they are all already written.
- Taking orders is the online ordering course for your own channel and the delivery course for third parties. If an app takes orders, it is a shell around what those two already taught you.
- Taking payment is the payment processing course. Nothing about a phone changes fees, chargebacks or settlement.
- Loyalty programmes are the loyalty course, and marketing them is the email and social courses. An app can carry a programme; it cannot design one.
- Bookings are the reservations course you just finished.
What is genuinely this course's own is the decision itself. Topic 2 is the one most readers should stop at: everything a customer's phone already does for you without a single line of code being written, which for the overwhelming majority of restaurants is the whole answer. Topics 3 to 6 are for the minority who get past topic 1 — how to choose, how not to be deleted in the first minute, how to earn a second visit, and how to know when to switch the thing off.