The correlation coefficient uses the =CORREL formula and calculates the relationship between the time spent (behavior) and retention rates to quantify their relationship. A value closer to 1 indicates a strong positive correlation, while a value closer to -1 indicates a strong negative correlation. In this case, .39 is a weak positive correlation! We can conclude that more time spent on the platform for an overall cohort of users that signed up will have a positive correlation with how many of those users perform the Aha Moment.

The color coding in the heat map represents the value of the quantitative variable: the behavioral metric. This allows me to visually compare it against a heat map of cohort retention rates. Higher values of time spent are shown in a warmer color (red), while lower time spent values are in a cooler color (e.g. blue).

This column marks the week when a group of users first interacted with the product.

Subsequent columns represent activity_week, showing the weeks’ behaviors when these cohorts were active, even if they did not perform the aha moment activity.

The intersection of a row and a column reflects the behavior: In this case, the total time spent on the app.

Crystal Widjaja

I used this Google Sheet to serve as a behavioral cohort retention tool to quickly visualize and analyze user behavior as heatmaps alongside retention metrics and calculate the strength of the correlation using =CORREL.

The correlation coefficient uses the =CORREL formula and calculates the relationship between the time spent (behavior) and retention rates to quantify their relationship.

The color coding in the heat map represents the value of the quantitative variable: the behavioral metric

This column marks the week when a group of users first interacted with the product

Subsequent columns represent activity_week, showing the weeks’ behaviors when these cohorts were active, even if they did not perform the aha moment activity

Behavioral Cohort Analysis at Graas

Cohort Builder - Quantitative

Sign up to see more