Recently, Lean Startup techniques provided great validated learning as well as some hard lessons for us. This was for an adjacent product in a company that already has a large customer base with existing recurring revenue. Zach Nies, from Rally Software, helped us us out and kindly allowed me to share several templates that are very helpful. The templates can be found here, courtesy of Zach.
The nature of RISK is very different in an established company than in a startup
The more you have, the more you have to lose. This is a big problem for larger organizations. Experimentation requires careful attention to safety, both for the organization and its customers. In the templates above, the one for experimentation helps you organize your experiments and also consider safety.
You will also find an Innovation Sandbox in the templates. This contains rationale and working agreements to provide the ability to work differently in an A3 strategy format. Based on the impact of risk in your organization, I recommend this includes specific financial, resourcing and communication protection. This is because the vision is not shared like in a startup. If we meet quarterly goals we agree on AND we didn't learn that we should stop doing what we are doing AND some external events, like Force Majeure, do not take place, then we should have a reasonable expectation that we shall continue on. Else, it will feel like justifying our existence every quarter.
Other templates help structure Empathy Interviews. These help frame the problems so we can really get to the essence of people's journeys. Also. I really appreciate UX Journey Maps. In the simple form we used cross-functional flow charts with how people are feeling at each step. This beings humanity back into the equation.
It is very important that we validate the problems exist AND that we believe we should solve them
Only then should we start validating solutions. One of the templates represents Leap of Faith Assumptions. These assumptions are the main ones that if not validated could render all other work useless. Over time, however, we found we can be working on more granular of problems and solutions at the same time. There was a time that we stopped experimenting under pressure to get a more coherent product. This was when the select few co-creators we worked with started making very specific requests and were consistent among the different groups. Experimentation can be very low cost at first. However, it can get very expensive and not yield learning as we progress. Not everything has to be validated through experimentation. This has to be constantly compared to the leap of faith assumptions.
Lean Canvas is a great way to track the evolution of a business model
Experimentation needs to be wrapped into a business model or the effort can wallow endlessly. I thought the organization of the Lean Canvas sections confusing until I internalized the "Product on the Left" and "Market on the Right" fill order forced a balanced view of problem, customer, value, solution, etc. Lean Canvas is a great way to track the evolution of the business model as progression happens. We did not update it all the time but in cycles, especially when we did a "Zoom In Pivot." Consider keeping leap of faith assumptions on your Lean Canvas, or other Business Model canvas. See Ash Maurya's site for more. Do not forget internal risks, which are missing in the Lean Canvas Risk Iteration Path.
It is much easier to add a bunch of people in a larger company. The "starvation quotient" of a startup is actually a positive in this case. At some point, scale will need to be there. This is where the resources of the company are great. In my experience, executive patience is a year to a year and a half for a major new effort. Do not get the early experimentation phase too well advertised unless you can keep within the "window of patience."
As you are starting to have success in a larger company, new risks show up
This is why companies carve off a group with its own management, spin out new companies and isolate the groups for protection. Geoffrey Moore describes it using a horizons model. The transition from experimentation to materiality (10% of revenue) is the hardest part. As Moore says in his book Escape Velocity, "Why would a salesperson work ten times harder for a sale one tenth the size for the new product rather than keep selling the existing product?" Take a look at Zach's model and consider where you are. It might help you navigate and explain your situation easier.
In spite of the obstacles to innovating in large organizations, this was one of my favorite ways to build software. We built things for two reasons:
- To learn
- We knew people wanted it and it solved problems
Given the need to innovate in an organization with existing revenue and customers will happen again, I expect I will use these techniques again.