If you’ve been involved with a startup in the past several years, chances are you’ve heard the phrase ‘minimum viable product’ quite a bit.

Thanks in part to prominent Silicon Valley figures like Eric Ries, the father of the Lean Startup, and Steven Blank, a successful serial entrepreneur, more and more young businesses, as well as product managers at larger, more established companies, are trying to perfect the art and science of the MVP.

But what is a minimum viable product, and how do you get there?

The MVP Defined

On paper, defining ‘minimum viable product’ doesn’t seem that difficult. Which is probably why many consider an MVP to be a product that is just barely commercially viable and nothing more. Makes sense, right?

But it’s not that easy, at least according to Ries. As he has explained, “the minimum viable product is that version of a new product which allows a team to collect the maximum amount of validated learning about customers with the least effort.” The crucial part: being able to collect data that will drive learning about customers. As Ries suggests, “In a lot of cases, this requires a lot of energy invested in talking to customers or metrics and analytics.”

Getting to MVP

If we accept that a meaningful MVP is about more than just minimum and a lot more about maximizing the amount of insight we can glean from potential customers with the least amount of effort, the question becomes: how do we get there?

Unfortunately, an MVP becomes more art than science when it comes to answering this. According to Ries, “It requires judgment to figure out, for any given context, what MVP makes sense.” The good news is that there are some common sense steps that are prerequisites for figuring it all out.

  • Think about it. If your ultimate goal with an MVP is to learn, and not to sell, give some thought to what you need to do. You may discover that the MVP you need looks a lot different than the MVP you thought you needed. More specifically, in some cases, you may find that your first attempt at creating an MVP doesn’t even require you to build something functional in the traditional sense.
  • Know your customers. Don’t let the phrase ‘minimum viable product’ fool you: the concept is decidedly more about the customer than the product. With this in mind, it becomes clear that to employ an MVP approach, knowing who your potential customers are, and having a way to get them involved, is obviously a prerequisite. If you’d rather make assumptions and not interact with customers, the MVP approach is not for you.
  • Establish KPIs. To maximize the amount we learn from customers using an MVP, it’s crucial to establish KPIs that you think will be important from the outset.
  • Identify the data needed to measure KPIs. With KPIs identified, ensure that there’s a means to collect the data necessary to measure those KPIs.
  • Collect more data than you need. The KPIs you’re looking at may or may not be the KPIs that allow you to maximize customer learning. So it’s smart to collect as much data as you can and be prepared to actually analyze it, looking for nuggets of customer wisdom you weren’t expecting.
  • Remember that MVP is a journey, not a destination. The first product you build may not be your MVP. Again, the MVP concept is all about customers and being able to learn from their use of your MVP. That almost always requires a significant amount of iteration.

Beyond startups

While the minimum viable product concept is probably most popular amongst startup entrepreneurs, there’s no reason that established businesses can’t put it to use. Whether you’re launching a new product or adding new features to an existing product, the general idea that you should try to maximize how much you learn about customers with the least amount of effort is one that can easily be applied to a variety of product development scenarios.