The Lean Startup concept is popular not only in Silicon Valley. It's used by thousands tech entrepreneurs around the world. At the Internews Center for Innovation & Learning, ongoing experimentation and innovation have been driven by lean startup principles.
Pilots featured on our sites are experimentation to the core. The purpose of the pilot is to test a hypothesis. Hypotheses are usually related to specific information needs that are identified with the help of research and on-the-ground experience. So the best way of getting results begins with understanding needs. There are several questions I would like to focus on to explain the design and implementation process.
1) Why pilots don't necessarily lead to projects
Most of pilots are short-term experiments designed to test theories with the flexibility of loose management. It's like a free flight. The measurements for success are undefined, which means the team doesn't feel pressure of dat-to-day management or to report on the process. Since the pilot is an exploration, it may not necessarily produce results that lead to a scaled-up project, but rather the results inform the design of future activities in a range of ways.
2) How to evaluate the work
Most of people who work in a media or tech office consider the day to have been productive if they have written a piece of code, completed a document or responded to all emails. But pilots contain a high degree of uncertainty for the team; the website might crash because the server was overloaded, partners might not like the direction of the pilot or the results; the target audience may not be responsive to the outreach campaign. A huge set of variables both anticipated and unanticipated might impact results. Success is measured instead by our success in documenting those variables and their impact on your hypothesis, both big and small. This documentation advances the pilot to its goal--producing a prototype of the system that could be scaled up to a project.
3) The core of the project--MVP
The minimum viable product is the core of any project. It boils down the the essential functions of the project necessary for basic operation.Those essential functions are determined by the hypothesis being tested. There should be two different experiments being conducted simultaneously: the hypothesis of value and hypothesis of growth. The former shows how important the service is to a target group and is conducted with the help of interviews and meetings with people that represent target audience. Hypothesis of growth answers the question of how many people may come and remain users of the service. It's based on assumptions and correlation of similar services.
It doesn't mean that MVP is simplistic. It simply identifies the method by which you can quickly and inexpensively set up an experiment to test a hypothesis that may inform future iterations. Many think of a successful start-ups as producing revolutionary technology that change the lives of millions of people. Modest ideas designed to test a limited range of functions can inform and attract funding for more ambitious prototypes down the line.
This small and unpretentious concept--the minimum working product--can be anything from a simple added feature on a website to a full working prototype of a new communications system.
The minimum viable product is dicataed by the fundamental assumptions that you want to test and your ideas about how they can be tested. But it is important not to overestimate the number of features that you want to include a minimum product. There is only one simple rule: if you are in doubt about some of the properties to include or exclude, then leave them out.
4) Hypothesis and evaluation
So the pilot is a process of testing ideas for specific information services. By carefully documenting the process and modifying the product according to set variables, the basis hypothesis may be proven. When the pilot starts to grow naturally and the MVP is constantly modified and tweaked for better results, then the experiment is ready to become a fully funded project.
For full experience please review lecture on Lean Startup approach to launching projects from Eric Ries the author of The Lean Starup.