At GroundControl, we’re a big fan of David J. Bland’s Testing Business Ideas book because it closely matches how we look at the journey from a great idea to a validated business model. It provides many examples of which experiments you can run to de-risk as much as possible. You could even say that GroundControl is Testing Business Ideas software.
In this article, I’d like to dive into how you can test your own ideas with GroundControl.
Assumptions Mapping is the process of determining “What is currently most risky?”. It uses the HiLo matrix by Dan Toma to map assumptions on two axes:
- How much do you know about this assumption? With Everything on the left and Nothing on the right.
- What is the impact on your business idea when you are wrong about this assumption? With Low Impact on the bottom and High Impact on top.
This gives you a 2-by-2 matrix where you can ignore everything on the bottom since it has Low Impact and is not worth focussing on. Your logo’s color is a classic example; You might know nothing about the color, but the impact is very low, so please ignore it.
In the top-left quadrant are all assumptions that you should discuss. High Impact assumptions where you know relatively a lot about. Why haven’t you validated or invalidated these assumptions yet? Do you need to run one more experiment?
And last but not least, the top-right quadrant with the risky assumptions. The closer to the top right, the more focus you should give these. Test these assumptions first.
In GroundControl, we use two sliders when adding assumptions and show only the Most Risky, Risky, and Discuss assumptions.
Stage Relevant Assumptions
Besides the Impact and Knowledge, we use a third characteristic of an assumption to determine what is currently most risky, and that is the stage the startup is currently in.
We learned that every startup goes to roughly the same four stages:
- Problem: Prove there is a problem (and who has that problem)
- Solution: Prove your solution can solve that problem
- Revenue: Prove your customer is willing to pay for your solution
- Scale: Prove you can scale your business idea from the early adopter to a bigger audience
For each stage, we look at what startups should focus on. For the Problem stage that is the Customer Segment and the Problem, for the Scale stage the Channels to reach customers and the Growth Engine.
You can create your own framework by picking the relevant building blocks from Alexander Osterwalder’s Business Model Canvas and arrange them according to the stages that are relevant for your organization.
This helps teams to understand that when they are trying to Prove the Problem, focussing on the Channels is a bit pre-mature.
Lean Startup-style Experimenting
After you have determined what assumptions are currently most risky, it’s time to find out how to test them.
Learn or Confirm
You can run two types of experiments:
- Learn experiments to learn more about a certain segment
- Confirm experiments to validate or invalide an assumption
You often first run a Learn experiment to create new assumptions and then a Confirm experiment to validate or invalidate these assumptions.
Create an Experiment
When adding a Learn Experiment you think about what you want to learn about the segment you selected, for example to learn which problems your customer is currently experiencing.
When creating a Confirm experiment you select the Assumption to (in)validated, you create a Hypothesis and think about ways to measure the success or failure of your Experiment.
Select an Experiment
Based on the segment you’d like to run your Experiment in, we suggest some of the most used Experiments. From Customer Interviews to doing a Picnic at the Graveyard.
We understand that your Experiment will slightly differ from our examples, so you can always adapt the Experiments or create your own from scratch.