Coming home from a client visit recently where an MVP took center stage in all the right ways, it occurred to me to examine how Confluence defines and views an MVP and why it is so misunderstood and misused in software development today.

First off, what is the official definition of MVP? While you can find differing definitions, I like the one proffered by Eric Ries, entrepreneur and author of Startup Lessons Learned, who popularized the concept. Mr. Ries states that “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.”  What do I like about this definition? I like it because it is really about the customer, not the product.

Often what we’ve seen is that product development organizations hear minimum and viable together and they think “easy”…as in, let’s build Product A because it is the simplest to build and the easiest to deploy. That kind of thinking is diametrically opposed to what the term really means.

At Confluence, we are careful to ensure that our validated learning comes from multiple sources. These early sources of learning are what we call “lighthouse clients”. Wikipedia uses lighthouse clients synonymously with early adopters. We think of them as being slightly different. Just because someone is an early adopter does not necessarily make them a good lighthouse client, but it is crucial that a company you choose as a lighthouse client be comfortable with being an early adopter.

Lighthouse clients work with our product development organization throughout early and frequent product iterations validating functionality and usability. The lighthouse client may not have actually committed to licensing an early release. The early adopter of our product will license the product when it is released. The lighthouse client is invaluable in helping us determine what scope of functionality actually makes our product viable for a particular market segment.

The client I referred to earlier was indeed a lighthouse client and they had been working closely with our product development organization on this particular MVP. Their ultimate validation came in the form of licensing and implementing the MVP before the official commercial release, otherwise defined by Ries as the Minimal Marketable Product (MMP).

I expect some readers are wondering if we made the mistake of developing a custom product based on the feedback of one client. While we certainly took some direction from them, other clients in the market segment were involved in the development and validation of this product. The result was the launch earlier this year of Unity® Publishing and based on discussions our product and sales teams are having, I am confident we have developed a true market solution.

Have you had any experience with MVPs, good or bad, as a developer or a client? We’d love to hear from you.