Minimum viable product (MVP). What is it and why should you care?

A minimum viable product is a great way of building user centric digital services in a fraction of the time. It will also lead to big cost savings.

I write a lot about the idea of a minimum viable product (MVP). But I don’t think I have ever taken the time to explain exactly what they are. If you have heard the term but are unsure what it means or why it matters, then this is the post for you.

What is a minimum viable product (MVP)?

To understand what a minimum viable product is you must first understand how most products are created. To demonstrate the process, let’s use a simplistic project for constructing a new building.

Management has an idea

The idea is formed

Step one is that somebody has the idea for a new building. This is often a senior manager somewhere. The idea might have been informed by some research into the market, but often it is not. Even when it is, the idea often lacks definition at this stage.

Plans are laid

Step two is a planning or specification phase. This is where all the details for the building are hammered out. This is an important stage, because the cost of getting things wrong would be high. Buildings aren’t cheap and the cost of changing things is high.

Organisations sometimes consult with the eventual occupants at this stage. But there is nothing to show them and so it is difficult to get a detailed sense of what they need.

Plans were drawn up
Construction begins

The project is implemented

Next, the plan is implemented. The building starts to be constructed. The focus at this stage is on delivering on time and within budget. The enemy is scope creep or unexpected developments. But if the planning stage was thorough enough these can be minimised.

The result is revealed

Once the build has been completed, the final building is revealed to the world. This is the crunch moment. Will the occupants like it? Will it have provided what they actually need, rather than what they said they wanted? Will they actually move in?

The building is complete

It is impossible to know the answers to these questions until the product goes to market. This is true even when you have done some research during its development. Worse still, if they don’t respond well there is no budget or time to make changes. The project only budgeted for some basic maintenance once the building was finished.

How a minimum viable product works

By contrast a minimum viable product allows you to gather reliable data about the customers reaction. Even better, there is no danger of over-engineering the solution. You can be confident you are only building features your audience will use.

Take the example of creating a public park.

Minimum Viable Product (MVP): A basic garden

Get to market fast

The initial park could be basic. A few trees and some grass. But as a result you can open it to the public early on. No long specification phase because the cost of implementation is low and we can always improve it overtime.

As the gardeners tend to the park each day they can talk to the public using the park. They can watch how they use the park and adapt accordingly.

Provides better feedback

For example they might get regular feedback that the park lacks colour. The gardeners might then decide to plant some flowers to brighten things up. After all the cost isn’t high and they were working on the garden anyway.

Minimum Viable Product (MVP): Flowers added to the garden
Minimum Viable Product (MVP): Pond added to the garden

Allows more informed research

Next, people start to say it would be nice to have a pond. This is a bigger investment. But a bit of research reveals that it would attract a new audience of model boat enthusiasts. With that knowledge the owners decide to make the investment.

Lets you observe natural behaviour

Finally, the gardeners observe that people are having to sit on the grass. That isn’t a problem for most, but some elderly people are struggling. They then decide to add a bench and afterwards notice that the elderly spend longer in the park.

Minimum Viable Product (MVP): Bench added to garden

Why minimum viable product offers a better way

As you can see a minimum viable product is a much more flexible way of working. It isn’t right in all circumstances. But for a digital project where the cost of change is low, it is worth considering. This is because it offers some distinct advantages:

  • It lowers the cost of development because you only build what people will use. Our park keepers could have built a skateboard park. But it turned out there was no demand for it so they didn’t make that investment.
  • It is faster to market because you can start with something basic and there is no need for a long specification stage. This gives you a competitive advantage when it comes to grabbing market share.
  • It leads to a more desirable product. This is because it gives people exactly what they need and they don’t end up paying for features that just clutter the experience.
  • This in turn leads to increased customer satisfaction and word of mouth recommendations.

Also in the case of digital products it leads to easier to use solutions. That doesn’t just increase satisfaction, it reduces support costs too.

Minimum Viable Product (MVP) Infographic
A minimum viable product isn’t a magic bullet. But when building a digital service it is often better than the alternative.

Don’t misunderstand me. A minimum viable product isn’t a magic bullet. It is not always the right approach. But when building a digital service it is often better than the alternative.

  • Estevez Atya

    The product is not the same.
    How would you make a building with MVP?
    Would you make big empty levels, without walls first?

    • You actually wouldn’t. That is kind of my point. A building has a huge cost of failure associated with it. It is not compatible with MVP. But equally it wouldn’t make sense to plan a park like planning a building. Unfortunately digital services are more like parks and yet managed like building projects.

  • Johan

    Nice summary, went in to the general Slack channel! But I do not agree with on one thing “… After all the cost isn’t high and they were working on the garden anyway.”. It’s more likely that the gardeners are working 10 parks and right now they are building a pond in one of the others.. The cost of changing back and forth between projects need to included, if we only do a MVP, when do we have to go back and improve, a week, a month, a year? Is it more cost-efficient to plant flowers directly? I do like to MVP-model, but the hard part it’s to integrate it.

    • You make a good point. But that is not a problem with MVP. That is a problem with how we manage people in a modern business.

      We still manage people like we are in the midst of the industrial revolution. We believe that if we push people harder and force them to be ever more efficient we will make more money. But that simply isn’t true with Knowledge Workers. It makes sense on a production line. But with digital in particular we need to allow time for experimentation, learning and evolution. That is because unlike a production line the final product is infinitely more complex and is influenced by an enormous number of variables.

      To be honest there could be a whole book on why we need to change our working practices to be more effective in the digital era. But sufficed to say the problem isn’t with MVP.

  • Alexandra

    Good example, Paul, it makes your idea very clear, this comparison of the difference between building a house and a garden. I agree that for digital projects MVP might be very useful as it would minimize expenditures related to modification of a product done without market testing. As Matt Evans explains in his article ( most of the ideas producers have
    about required product/service features turn out to be wrong, so the feedback from early adapters is very important.
    Also, I would like to draw attention of those who outsource projects like web or mobile app development to consideration of MVP creation with your prospect vendors. Though, some companies clearly mention this step (, and I find it very helpful for those who are willing to develop an app and are new to this.
    Just to add little inspiration, examples of big names that used MVP (or some kind of prior market testing) to build their final products:

  • Kael Konvard

    Awesome article!

  • Kami Lay

    It`s a nice article, thank you!I think it would be interesting to find out some more info about what the use of MVP is. I would advice a useful article.
    I read yesterday.

  • Matt0_0

    This is a really great explanation and nicely put together. The issue I’m struggling with is how to apply the MVP model to a website redesign when a client already has a feature rich website and are adamant about not loosing functionality but don’t have the budget. How you define or decide what should be within or outside scope of the MVP? I can see how this is a great approach for new build projects, but perhaps it’s not ideal for a redesign project?