A car wash app is a membership system that happens to open a gate.

The unlimited plan is the product

Start with the arithmetic, because it sets the price and it explains the panic. An unlimited plan is sold on frequency and paid for by the members who do not turn up. Somebody who washes twice a month, on a plan priced near three single washes, is comfortable. Somebody who washes four times a week is not, once you count chemical, water, power and the space in the queue. The price is a bet on the average, and the app is what tells you whether the bet still holds.

So the number worth a screen is not sign ups. It is washes per member per month, split by plan tier and by site. A car wash membership app that reports revenue and headcount looks healthy right up to the month the line gets long and the walk up traffic stops fitting through.

The record behind a member is small and specific. Plan tier, the vehicle it covers, the start date, the next billing date, the current status, and the reason for the last change: card declined, paused for winter, cancelled, refunded. Those states are most of what a membership management app is. An app that stores one boolean called active will be lying to somebody within a month.

Cancellation is a legal duty, not a courtesy. For a recurring charge sold online in the United States, the seller has to disclose all material terms clearly and conspicuously before taking billing information, obtain express informed consent before charging the card, and provide a simple mechanism for stopping the charges. A plan that starts with two taps and stops only by phoning the site on a Tuesday afternoon fails the third one.

15 U.S.C. 8403, negative option marketing on the internet

The gate has seconds to recognise the car

A drive through lane moves. Whatever proves membership at the entry has to finish before the car in front pulls forward, with the phone in a cupholder, in the rain, and on many sites with no usable mobile signal inside a concrete tunnel. That one constraint removes most of the elegant ideas.

Three approaches survive a queue. A windscreen tag the reader picks up with no action from the driver. A plate read, where a camera matches the vehicle to a plan. A code on the phone, cheapest to build and slowest in use, because it costs a wake, an unlock and a window down. The tag and the plate also enforce the rule that plans are sold per vehicle. A code enforces nothing, because a phone can be handed out of a window.

So the phone app is usually not the thing at the gate, and treating it as a scanner is the common wrong turn. It is where the plan is bought, paused, upgraded and cancelled, where receipts live and where a second vehicle gets added. Whatever does read the car has to keep working when the link to billing is down. Letting a vehicle the site already knows through, and reconciling afterwards, costs less than a barrier that refuses a paying member on a Saturday.

Who takes the money for a wash, and on which rail

This question decides the build, and the answer is already written down. Apple's App Review Guidelines split what you unlock inside an app from what the customer consumes outside it. Unlocking features or functionality within the app, subscriptions and premium content included, has to use in-app purchase. Goods and services consumed outside the app do not.

A wash is consumed outside the app. Guideline 3.1.3 covers this case directly: where an app lets people buy physical goods or services that will be consumed outside of it, the payment has to be collected some other way, and Apple names Apple Pay or ordinary credit card entry as the examples. An unlimited plan is a recurring charge for a physical service, so it belongs on your own payment processor, and the commission Apple takes on in-app purchases is not part of its price.

Google draws the same line. The Play Console payments policy does not require Google Play's billing system where the payment is primarily for a physical service, and its own list of examples includes gym memberships. A monthly wash plan sits in that bracket. Read the live wording before you launch, because these policies get revised.

The practical consequence is that you need a payment processor with a real subscription product behind the app: a card on file, retries on a declined charge, proration on an upgrade, and a message to a human when it all fails. Newly does not ship built-in payments, so that integration is a decision you make and wire up yourself.

Apple App Review Guidelines, 3.1.3 on goods and services outside of the app

A detail is an appointment, an express wash is a queue

The moment a site sells something that takes an hour, the software changes shape. A tunnel has throughput. A detail bay has a calendar. A detailing booking app has to know how long each service takes, how many bays and staff can run at once, which services cannot share a bay, and how long a slot is held before it is released again.

The rules that cause trouble are the dull ones. A ceramic coating that keeps the car overnight. A van that does not fit in bay two. A last appointment that has to finish before closing, so its start time moves with the length of the service instead of sitting on the hour. Those are the constraints of any other booking app for business, with vehicle size added as an extra axis.

No shows are where the money goes. A deposit at booking is the usual answer, and since it pays for a service consumed outside the app it follows the same payment rule as the plan. A reminder the evening before costs nothing and saves more slots than most operators expect.

A car wash loyalty app sits underneath all of it and is far simpler: count the visits, reward the count, pay nothing in a quiet month. For a site without a tunnel that often beats a plan outright, because somebody who washes once a month will never agree to pay every month.