MVP: The problem with failing fast

Born out of the startup movement in Silicon Valley, Minimum Viable Products (MVP’s) have quickly gained popularity, spreading to both small companies and larger enterprises across numerous industries, and understandably so — a 90% failure rate for new initiatives is too expensive for everyone involved.

Historically, companies would often place large bets on untested new ideas — an approach that lost money because initial guesses about customers needs are often wrong.

In order to address this, modern businesses have shifted to a model in which they frame initial business ideas as customer experiments, rather than a dogmatic vision of what the business is about. The heart of this approach has been dubbed the Minimum Viable Product (MVP). MVP’s are designed to allow busineses to quickly and cheaply validate their business ideas with real customers — this can result in new businesses changing direction entirely or larger enterprises killing initiatives quickly. This has given rise to the popular but flawed concept of Failing Fast.

The problem with failing fast is it frames MVP’s as cheap customer releases that can often fail. This isn’t the case; MVP’s aren’t releases, they are experiments. The distinction is important for two reasons: Firstly, it avoids poorly funded, low quality releases to customers.

Secondly, it leads to the important idea that MVP’s are always successful if they prove or disprove a hypothesis about customer needs. MVP’s only only fail if they fail to test a valid hypothesis, cheaply and quickly.

One of my favourite examples that underlines the points above, comes from the story of Nick Swinmurn, an entrepreneur who, in 1999, asked himself whether people would buy shoes online, or whether they would always need to try shoes on first. In his own words:

“One day I was at the mall and couldn’t find a pair of the Airwalk desert boots I wanted. So I thought, Why not do an online shoe store? I went to Footwear Etc. in Sunnyvale [Calif.] and said, “I’ll take some pictures, put your shoes online, and if people buy them, I’ll buy them from you at full price.” The store said okay, and I got a few orders. Then I went to a shoe show and thought, I need to put this giant collection online.”
Interview with Nick Swinmurn, Fortune.com

The seeds Nick sowed evolved into a business that would later sell to Amazon for $850 million dollars.

Nick’s MVP was cheap and fast because it wasn’t burdened with the need to last beyond proving the market for online shoes. Nor did it need to establish a brand that set future expectations for customers. Once the experiment was over, the site could be torn down, which is exactly what happened — Nick’s “ShoeSite.com” disappeared, and gave way to Zappos.com.

MVP’s are useful, but they’re not about failing fast, they are about learning fast and at low cost. If you’re planning on using MVP’s, this distinction can help to shift focus on running more experiments on a limited budget, and increasing the odds of product success.

Leave a comment below, I’d love to hear from you.

Raj