If your internal innovation process is maturing, you understand that you need to start making decisions based on data rather than on gut feelings or some pretty pictures.
This ensures that you make innovation manageable, measurable, and visible for everyone.
To make this happen, you need to understand how to evaluate your innovation process by plotting stage relevant assumptions as well as how to make decisions based on real data. Here's how Esther Gons, co-author of The Corporate Startup and the forthcoming Innovation Accounting, explains how.
A snippet from Esther Gons’s session during the Innov8rs Connect Unconference, June-September 2020. To watch the full session recording, join Innov8rs Community with a Content or Premium Pass.
Innovation Lab and Innovation Accounting: How To Set it Up
For disruptive innovation, you need a different system. A system that does not contribute to your current business models or operations, but rather a second system that allows you to find new business models in the future. This is called a startup rule book, and you need it if you want to set up an ambidextrous organization. That second system also needs innovation accounting so that you can replace all the processes and KPIs from the current system, which are hurdles for your disruptive innovation.
To set up this system, you need to consider that there are multiple layers to data-driven innovation accounting:
- Your teams (how they are doing)
- Accumulated data from all of your teams (can you help the teams do better, what do you need to ensure that the funnel is filled with good projects)
- The whole strategic ecosystem (does this contribute to your strategic goals within the organization)
Systematic Approach to Innovation
It is essential to understand that all of this starts with your projects or startups, that’s where all the data will be coming from. By getting their data, you can figure out:
- How are they doing on a weekly or monthly basis?
- Can you help them?
- Are there any hurdles preventing them from learning with evidence?
That is what you need when you do data-driven decision-making, actual evidence of what your initiatives are learning
To accomplish these goals, the first thing you want is some kind of talking point with real data to understand how they are doing on a continuous basis. Not just every year or quarter. You want to know on a deeper level within a startup: what they are learning, how they are learning, and what experiments they are doing to provide you evidence, so that you can get the data you need to make decisions. Also, you want the teams to approach this in the same way so that you can compare them.
This data is always connected to the experiments that your initiatives are doing, and it is backed up with evidence that is coming out from these experiments. Even though the evidence may appear fuzzy in the beginning and data-driven at the end, all of it will help shape your decisions.
Once you have the first layer down, you can see how your teams are doing, if they are performing well, if you can see all the teams and where they are in the different stages. In the level above that, you will also want to see how your portfolio is doing. Are the startups proceeding fast enough to the various stages, how much have you invested, how much time did you spend in these different stages, and do you have an invalidation rate?
The Framework
The next big question is, how do you set up a system that helps you make these data decisions? First, you need a framework. It seems quite logical that you need a stage gate or framework to make decisions on. However, many companies end up creating such a framework taken from their current system.
Most traditional stage gates models from product development are borrowed from the current system. These models were contrived from twelve stage gates because they needed to decide based on actual strong criteria and risks.
Most of these stage gates are borrowed from the IT department or the digital department where, apart from some sort of discovery phase, the framework is based on a non-risk rollout that will give you some indication of how long the rollout will take.
Using existing stage gate frameworks for innovation makes data-driven decisions for disruptive innovation really hard. The criteria are still KPI’s, based on what the current business wants to see and based on business cases and models, which means you still have to guess or project.
Instead, Esther suggests you take a conscious effort in thinking about how different this stage gate association is if you look at startups and apply those learnings to your stage gate framework.
It is crucial to understand that most startups fail. Based on this mindset, you need a lot of initiatives to have some results in the future, and that can take a couple of years.
So, for example, if you want two initiatives to be successful, you need a lot of initiatives from the beginning. That is why stage gates are so important in setting up data-driven decision-making, and if that is not done right with the right mindset, then all the other things relating from that will be difficult as well.
To succeed, you need a VC mindset for investing because we are looking at high risk, high return. As a result, it is not a controlled rollout. Instead, it is about betting on many small bets to have some developments in the future. That is why you need to start with a lot of initiatives to have potential later on. Making sure to save for a few bigger bets of those that prove their potential along the way.