via the Game Thinking Academy, https://www.youtube.com/watch?v=C6rDUmpCO-o
5 steps to the scientific method:
- Hypothesis
- Subjects
- Experiment
- Data
- Conclusions
Write a testable hypothesis
Have clear, concrete assumptions
Take a skeptical view of your hypothesis–you’re looking for a market truth, not to prove out a previously held narrative of what the product should be
Choose the right subjects
Talk to the specific folks who will be the first ones to use your product–they might be a subset of your eventual audience, but start with them as your “beachhead” on the way to your destination
Run the right experiments
You need to run experiments that surface what people think of your product, not necessarily to succeed
Depending on where you are in the product design process, it might be:
- A concept scenario
- A diary study
- Low fidelity prototypes
- Full blown A/B beta testing
Analyze the data
You need to start with clear actionable data
Look for patterns that arise such as:
- Existing features
- Feature system preferences
- Privacy concerns
- Motivational triggers
Qualitative in-person testing with a few well-targeted potential audience members can help validate your evolving designs
As the design becomes more mature, switch to more quantitative methods
Draw conclusions, don’t rubber-stamp assumptions
Listen to the data-some things will plain fail, some things won’t be quite right but will lead you in a better direction, and some things will work pretty well
If everything tests well from the beginning, are your tests really robust enough to uncover the truth?