Stop abusing A/B testing, toxic experimentation culture, how to run A/B tests with rigor - Che Sharma - The Data Scientist Show #071
The Data Scientist Show - Daliana Liu
00:00
Using Sequential Testing to Solve the Peaking Problem in Experiment Programs
To ensure a healthy experiment, it's important not to jump to conclusions too early. Many experiment programs use T tests or Bayesian methods, which promise not to look at the results until the end. However, it can be challenging for companies to resist checking results, especially when there's a significant drop in revenue. Sequential testing is recommended as a solution to the peaking problem, as it allows for more flexibility in analyzing data.
Transcript
Play full episode
