What Is a Minimum Viable Product, and What Should It Leave Out?

A minimum viable product is the smallest version of a product built to prove the single riskiest assumption in a business idea. Everything that does not test that assumption waits, including payments when the real question is whether anyone shows up. The one thing an app MVP cannot leave out is the safety work the app stores require.

What is an MVP, really?

An MVP is a test that happens to be real software. In business terms the meaning is simple: "viable" means real people can use it for real, and "minimum" means you build only what the test needs. Treating it as a cheap first draft of the finished product is where most MVP budgets go wrong.

Most founders get the idea right and the scope wrong. They add what the finished business will need: payments, admin dashboards, merchant tools, loyalty points. Each feature feels responsible, and each one rests on an assumption nobody has tested yet.

The fix is to write down the one thing that must be true for the business to work, then ask of every feature whether it helps prove that thing. If it does not, it goes on the list for version two.

Why payments are often the wrong thing to build first

For a product with two sides, such as a marketplace, a directory or a social network, the riskiest assumption is the audience, not the payment. Merchants will not pay to reach people who are not there, and a payment system in an empty marketplace tests nothing. Guides written for shops, where taking money on day one is the test, miss this. It is also why, writing about how a trades directory should charge its providers, we said adverts and paid placement need traffic first.

Running out of money is usually the symptom. In CB Insights' analysis of 431 VC-backed startups that shut down since 2023, 70% ran out of capital, but the firm says that is almost always the final cause of death rather than the root problem. Poor product-market fit appeared in 43% of the failures. Building the revenue machine before you know the product fits spends capital on the wrong thing.

When we scoped a location-based social app for a South African founder, the long-term plan was local merchants paying to reach people nearby. That only works if a 2km radius holds enough people posting and watching each other's content to be worth selling to. So we recommended a first build that cannot take money at all: no in-app purchase, no merchant accounts, no paid placement. The proposal calls that a decision, not an omission, and names the cost up front: the founder funds the runway until revenue ships.

What should a minimum viable product include?

An app MVP must include the safety and account features the app stores require, even though no user will ever ask for them. For an app where people post content others can see, Apple's App Review Guideline 1.2 lists four requirements: filtering objectionable material before it is posted, a way to report offensive content with timely responses, blocking abusive users, and published contact details. Apple has also required in-app account deletion since June 2022 for apps that let people sign up.

In that proposal, blocking, account deletion, content reporting, a terms acceptance step and store submission checks came to 87 of 423 hours, roughly a fifth of the build. Even then it met three of Apple's four items, because posts are moderated after a report rather than filtered first, and we flagged that as the likeliest friction at App Review.

Before you brief a developer, finish this sentence: "This build proves that..." Then ask which features in the quote do not help prove it, and how many hours of store and legal work it carries, a line worth finding when you read an app development quote. Ask what gets added next, and whether the architecture takes it without a rebuild. These are the questions we start with in our MVP development work.

If your product needs other people to show up before anyone will pay, build for the audience first and design the payment in, ready to switch on. Then the choice between paid placement, subscriptions and an ad-supported model gets made with real usage behind it. If you sell directly to the person using the product, the payment is the test, and it belongs in version one. Either way, budget for the work no user sees, because an app without it does not reach the store.

Questions about minimum viable products

What is the purpose of an MVP?

The purpose of an MVP is to prove the riskiest assumption in a business before you spend on the rest. For a marketplace or social app, that is usually whether people will show up. When we scoped a location-based social app, we recommended a first build that could not take money at all, so it would test the audience alone.

What does MVP mean in business?

In business, MVP stands for minimum viable product: the smallest working version of a product that real customers can use, built to test whether the idea holds. It is a learning tool rather than a cut-price launch. The scope comes from one question, what must be true for this business to work, and everything else waits.

What is an example of a minimum viable product?

A location-based social app we scoped for a South African founder is one example. The long-term plan was merchants paying to reach nearby users, but the first build we proposed left out payments, merchant profiles, chat and push notifications. It kept only what tests whether people within 2km will post and watch, plus the store safety work, minus pre-publication filtering.

When should you use a minimum viable product?

Use a minimum viable product when you have a clear idea, a limited budget and an assumption you cannot confirm without real users. It suits new products, marketplaces and apps entering an untested market. If the risk is purely technical, a prototype answers it more cheaply. If demand is already proven, build the fuller product instead.

What are the problems with a minimum viable product?

The main problems are building too much, which buries the answer under features, and building too little, which gets the app rejected or ignored. Missing features also distort results. We left push notifications out of one app MVP and told the founder that early engagement figures should be read as a floor, not a true reading.

Does a small business in South Africa need an MVP?

Yes, if you are about to spend real money on software built around an idea nobody has tested. An MVP lets a South African business owner test demand before committing the full budget. It does not need to be an app: for many service businesses a landing page, a WhatsApp flow or a simple booking form is enough.