The key to a successful MVP is not any specific feature or any product-oriented characteristic. In fact, a successful MVP will almost always not be very good.

Purpose of an MVP

The purpose of an MVP is not to design something and then test it to see if it's good. The purpose of an MVP is to conduct as quick of an experiment as possible in order to learn something about the ultimate direction your product should take to be successful in the market. You use an MVP for the same purpose that you conduct a scientific experiment: to test whether or not some kind of theory holds true in the real world. But you have to be careful because releasing any product that's not very well-developed can always do damage to your business. Therefore, the MVP must be able to test an idea in a way that's very small in scale and limits the negative impact that can result from releasing an experimental product.

By definition, this will also be fast, and an MVP must be fast in order to allow you to learn at the same pace as the evolving needs and conditions of your market. So you must be able to conduct the MVP test quickly, and in a way that's designed to harvest feedback quickly so you can evolve from the MVP into something that is a better product and will make more money, although that will entail more effort, cost, and time to do.

The reason we use an MVP is because, in reality, it is pretty much impossible to truly know what will constitute a good product and how that product will fit with the market. Markets are simply too organic and chaotic to really be able to do this in a predictive way. In theory, it is often the job of someone like a Product Owner or a Product Manager to know what type of product to build and how it relates to its market. In reality, product people are usually not very good at this. Nobody is. Not many of them will admit it, but it's true. So a good product person will use a strategy to learn what markets want, and what kind of product to build. This is why we use an MVP: to collect feedback from the wild which allows us to learn what will work in the wild.

So, the key characteristics of a successful MVP will be that:

• It is small in scale, fast and cheap.
• It doesn't create too much negative impact when you release the half-baked idea into the market to see how it behaves.
• It is designed to gauge whether a product is viable at a larger scale. This one is tricky because often you have to detect very weak signals to learn whether this will be true. So a successful MVP is designed with an understanding of what those are weak signals might be and how to detect them.
• It yields useful lessons about what users and markets actually want related to your idea or product so that you understand how to improve it and scale it up.
• Finally, one very important characteristic of an MVP is that it will just be one of many MVPs you do. If an MVP takes two years to build and release, it isn't an MVP. It is a traditional product launch with lots of front-end cost and risk. You should always be doing a lot of small-scale testing so that you can always be learning what your product(s) should really look like. So if an MVP isn't one of many, it probably won't be successful — not because of a problem with the MVP itself, but because you don't understand the purpose of the MVP and you aren't doing what it takes to keep up with the market, and therefore the business itself is not likely to succeed.

Want to learn more? Take a look at our Lean Product Management and Development course.