You've seen how to test claims with A/B tests and spot patterns with time series analysis. Now let's practice deciding which tool to use and how to set up your analysis correctly.
Engagement Message
Ready to test your skills?
Type
Fill in The Blanks
Markdown With Blanks
Let's frame the hypotheses for an A/B test on a new "Add to Cart" button color.
Null Hypothesis: The new button color has [[blank:no effect]] on the click-through rate.
Alternative Hypothesis: The new button color [[blank:increases]] the click-through rate.
This structure helps us test for a [[blank:specific]] improvement.
Suggested Answers
- no effect
- increases
- specific
- decreases
- generic
Type
Swipe Left or Right
Practice Question
Is the outcome statistically significant (a real effect was found) or also practically significant (the effect is large enough to matter for the business)?
Labels
- Left Label: Statistical Only
- Right Label: Both
Left Label Items
- A new ad increases clicks by 0.01%, but costs $50,000 to run.
- Changing a font improves reading speed by 0.2 seconds on a 10-page document.
- A new server reduces page load time by 5 milliseconds.
Right Label Items
- A new checkout flow increases completed orders by 15%.
- A training program reduces employee errors by 40%.
- A loyalty program improves customer retention by 25%.
