4 Principles for Making Experimentation Count

Product Experimentation Should be Hypothesis-, not Feature-driven

Defining the Proper ‘Exposed Population’ is Paramount

Understanding Power is Essential

Three suggestions:

Failure is an Opportunity: Use it

Experiments do not fail- hypotheses are proven wrong

Conclusion

Airbnb Engineering & Data Science

Creative engineers and data scientists building a world where you can belong anywhere. http://airbnb.io

Thanks to Tim Rathschmidt and Adam Neary.

lindsay m pettingill

Written by

Data Scientist @Airbnb. I make data useful. I also like mapping. Madcap and meticulous.

Airbnb Engineering & Data Science

Creative engineers and data scientists building a world where you can belong anywhere. http://airbnb.io