How to calculate sign-in rate using the formula in Mixpanel?

Hi event, Please stand by as I search the Mixpanel knowledge base for an AI-powered answer. I normally take about 15 seconds or so. In case AI cannot answer your question, feel free to reach out to Mixpanel support at mixpanel.com/get-support. If you have follow-up questions for this thread, please tag me again with your question: @kapa.ai <follow-up question> 👋

To calculate a sign-in rate using formulas in Mixpanel, you can use the Formula feature in the Insights report. Here's how to do it: 1. Create a new Insights report 2. Add two metrics: - First metric: The "Sign In" event (or whatever you've named your sign-in event) - Second metric: An event that represents your total user base (could be "App Open", "Page View", or any event that all users perform) 3. Click the Metrics button, then select "Formula" 4. Use the letters assigned to your metrics to create a formula that calculates the rateFor example, if your sign-in event is labeled as "A" and your total users event is labeled as "B", your formula would be: LBKS(A/B)**100LBKSThis will give you the percentage of users who signed in out of your total user base.You can also multiply by 100 to display the result as a percentage rather than a decimal.Mixpanel supports several operators in formulas: - + (Add) - - (Subtract) - ** (Multiply) - / (Divide) - () (Parentheses to control order of operations)If you're on an Enterprise or Growth plan, you can save this formula for reuse across your organization. Free users can create formulas within reports but cannot save them for reuse.Mixpanel Insights DocumentationRemember that you can rename your formula by selecting "Rename" in the overflow menu of the formula for better clarity in your reports.

ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.

Let me check the knowledge base for you again. One moment! 📚

Yes, you can calculate this sign-in rate in Mixpanel using a combination of Aggregate Property per User and custom formulas. Here's how to approach it: 1. First, you'll need to track two key pieces of information: - Sign-in events with timestamps - A user property that captures when each user first signed in2. In Insights, you can set up your calculation: - Select your sign-in event - Choose "Aggregate Property per User" as your measurement - Select "Sum" to get the total number of sign-ins per user - For the aggregation option, select "Average" to get the average across all users3. To account for the years since first sign-in, you have two options: Option 1: Using Custom Properties You can create a custom property that calculates the years since first sign-in for each user, then use this in your formula.Mixpanel Custom Properties Documentation allows you to calculate new properties on the fly.Option 2: Using Formulas with existing metrics If you already have the total sign-ins and the time since first sign-in (in years) as properties, you can use Mixpanel's formula feature to create the calculation: LBKS(Total Sign-ins per User) / (Years Since First Sign-in)LBKSThen average this across all users.The "Aggregate Property per User" feature is particularly useful here as it allows you to: 1. Calculate values per user first 2. Then apply aggregations like average, median, or percentilesMixpanel Insights Documentation provides more details on these measurement options.For the most accurate analysis, you might want to filter out users who signed up very recently (less than a month ago) as their rate would be artificially inflated when converted to an annual rate.

ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.

Let me check the knowledge base for you again. One moment! 📚

Both definitions of sign-in rate are valid but serve different analytical purposes in Mixpanel. Let me explain the differences and when each might be more appropriate:

(Total sign-ins ÷ years since first sign-in, averaged across users)This metric: - Accounts for user tenure, normalizing activity over time - Helps identify long-term engagement patterns - Controls for the varying lengths of time users have been on your platform - Better reflects user stickiness over extended periods

(Number of sign-ins ÷ number of registered users)This metric: - Provides a straightforward snapshot of current engagement - Is similar to how DAU/MAU stickiness is calculated in Mixpanel - Easier to calculate and interpret - Better for short-term analysis and quick health checks

The first definition (average sign-ins per year) is generally better for fair comparisons across different user cohorts or time periods because it normalizes for user tenure. As noted in Mixpanel's blog about stickiness metrics: "Stickiness attempts to capture the frequency with which your users are compelled to use your app... it's specifically about how frequently your users open the app, engage with it, leave, and then come back for more on a different day." What is your stickiness metrics really saying?For comparing user engagement across different cohorts (e.g., users who joined in different years), the first definition provides a more equitable comparison by accounting for how long each user has been active.However, if you're looking to measure current engagement levels or compare against industry benchmarks (which often use the simpler DAU/MAU ratio), the second definition might be more appropriate.As Arjun Sethi from Tribe Capital notes in Mixpanel's blog: "The best approach is to compare against yourself over time. You want to increase your retention rate over time as you improve your product." Chasing (and finding) product-market fitIn Mixpanel, you can implement either approach using the Formulas feature in Insights, with the first definition requiring more complex property calculations to account for user tenure.

ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.