The role of a startup is to propose a new product or service that meets a consumer demand in a new way. This we know. However, what some forget is that the success of a startup depends on two unknowns: whether this need is real, and whether the product or service actually addresses it.

Answering these questions is far from simple. Startups often fail to understand one or both of these factors. And in the race to impress investors, customers and teams, they do not always have the resources or the willingness to make mistakes. But without trying new approaches, or testing their proposition, they don't learn enough about their present and future relevance. By the time the product or service is launched on the market, it is often too late to fix fundamental flaws.

“Everything you do in a startup is already an experiment” said Eric Ries, in a recent conversation with Google for Entrepreneurs. "What defines being an entrepreneur, or defines a startup, is that we dont yet know whats going to work and therefore have to go through a process of discovery."

Ries’ key point is that to produce the best product or service for customers, entrepreneurs have to develop them through testing. To avoid wasting a startup’s limited resources, and to achieve the design of the perfect product as quickly as possible, this means thoroughly planning and evaluating as they go. Ries calls this the principle of the “lean startup”.

In his book The Lean Startup: How Constant Innovation Creates Radically Successful Businesses, Ries proposes a new unit to measure progress, which he calls “validated learning”. The emphasis is on applying a rigorous process to product development, based on a “build - measure - learn feedback loop” that continues until the product has been honed to its ultimate state. “Lean isn't just about failing fast, failing cheap. It is about putting a process, a methodology around the development of a product” he says.

Underpinning this concept of experimentation is the principle of design innovation. In the same conversation with Google for Entrepreneurs, CEO of IDEO Tim Brown, who I interviewed last year in the Huffington Post, agrees with Ries. Too often, he argues, startups pose questions that they already know the answer to - or they choose experiments that will support an idea they have already conceived.

By contrast, Brown says: “One of the things that I think is particularly precious … is a willingness to think about the future in a big and open-minded way, and an ability to have confidence that an unknown outcome is feasible - and that a willingness to strive for an unknown outcome is ok.” In fact, both Brown and Ries are adamant that such uncertainty, when approached through design, is the only path to real innovation.

Ultimately, freedom to fail early and often is imperative for any business, big or small. The Lean Startup proves that experimentation provides value to any organisation that wishes to remain relevant in a constantly changing world.