Cookie Cats Game AB Testing

This project analyzes an A/B test conducted in the Cookie Cats mobile game to determine whether moving a key game feature (the gate) from level 30 to level 40 impacts user retention and engagement. Using Python, data analysis, and statistical testing, the study finds no significant difference between the two groups. The key skills demonstrated include data wrangling, exploratory data analysis, hypothesis testing, and data visualization. The business recommendation is to keep the gate at level 30, as moving it to level 40 does not significantly improve user retention or engagement.

  • Setting the stage

    To determine the better user experience, we conducted an A/B test, splitting our audience into two groups: Group A (gate30) and Group B (gate40).

  • Formulating Hypotheses

    The hypothesis is that retention is higher when the gate is at level 30 than when it is at level 40.

  • Data Exploration

    We begin by exploring our data, ensuring it's clean and understanding its basic characteristics through descriptive statistics and visualizations

  • Visualizing Data Distributions

    Histograms and boxplots reveal the distribution and spread of our metric for both control and test groups, helping us visually inspect any potential differences.

  • Conducting the T-Test

    Using a t-test, we compare the means of the two groups. The t-statistic and p-value inform us about the statistical significance of our results.

  • Results Interpretation

    The p-value indicates whether we reject or fail to reject the null hypothesis. In this case:

    Test Type: Non-Parametric

    AB Hypothesis: Fail to Reject H0

    p-value: 0.0509

  • Drawing Conclusions

    If we want to keep retention high we should not move the gate from level 30 to level 40.

  • Next Steps

    Modifying to include number of game rounds played or how much in-game purchases are made by the two AB-groups.

Previous
Previous

Chat with BigQuery

Next
Next

Customer Segmentation