Generating a Report for User ID Counts in Identity Clusters

Hi Lois, 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> 👋

Based on the knowledge sources provided, there isn't specific information about creating a report that directly returns the number of IDs in a user's identity cluster. However, I can share some relevant information about identity clusters in Mixpanel: 1. For projects using the Original ID Merge API, there is a limit of 500 distinct IDs per identity cluster. When this limit is reached, new distinct IDs can no longer be merged into the same cluster, which can lead to orphaned or duplicate users in Mixpanel.2. The User Profile Activity Feed report allows you to examine the Identity Cluster to see all the Distinct IDs linked to a user. As mentioned in the documentation: "The User Profile Activity Feed report contains all the information about a specific user that you are tracking. Using this report you can examine the Identity Cluster to see all the Distinct IDs linked to a user" Check Your Implementation.3. For projects using Simplified ID Merge, there is no limit on the number of identifiers allowed in an ID cluster, unlike Original ID Merge.While you can view the identity cluster for a specific user in the User Profile Activity Feed, the knowledge sources don't mention a way to create a report that specifically counts the number of IDs in identity clusters across users. You might need to use the User Profile Activity Feed to manually check this information for individual users.

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 could potentially use JQL (JavaScript Query Language) to analyze identity clusters, though there isn't a direct built-in function specifically for counting IDs in identity clusters mentioned in the knowledge sources.JQL is Mixpanel's custom query language that allows you to create custom queries using JavaScript code. However, it's worth noting that according to the documentation, "JQL is in maintenance mode" and "Mixpanel is not investing in JQL any further" JQL: Create custom queries using Javascript code.For your specific use case of counting IDs in identity clusters, you might need to create a custom JQL query. The approach would likely involve: 1. Using the People() function to retrieve user profiles 2. Potentially using the join() function which returns objects with user information 3. Creating custom logic to count distinct IDs associated with each userUnfortunately, the knowledge sources don't provide a specific example of how to query identity cluster information using JQL. If you need to perform this analysis, you might consider using Mixpanel's Data Pipelines add-on to export your events to your data warehouse for more flexible querying, as this is the recommended alternative to JQL for advanced custom queries.If you decide to pursue the JQL approach, you would need to write a custom JavaScript function that implements the logic for counting IDs in identity clusters, but the specific implementation details aren't covered in the provided knowledge sources.

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! 📚

I'll suggest an example JQL query that might help you count the number of IDs in a user's identity cluster. However, I should note that the knowledge sources don't provide a direct example for this specific use case, so this is an attempt based on the JQL functionality described in the documentation.LBKSfunction main() { // Join events with user profiles return join( Events({ from_date: "2023-01-01", to_date: "2023-12-31" }), People() ) // Group by user ID to get unique users .groupByUser(mixpanel.reducer.null()) // Process each user to extract identity information .map(function(user) { // This is where you would need to access identity cluster information // The exact implementation depends on how identity data is structured // This is a placeholder for the actual logic return { distinct_id: user.key[0], // You would need to implement logic to count IDs in the cluster id_count: 1 // Placeholder value }; }); }LBKSThis is a starting point, but there are limitations: 1. The knowledge sources don't explicitly show how to access the full identity cluster data through JQL 2. JQL is in maintenance mode as noted in JQL: Create custom queries using Javascript code 3. The join() function returns objects with user information, but the exact structure for accessing identity clusters isn't detailed in the provided sourcesFor a more reliable solution, you might consider: 1. Using the User Profile Activity Feed UI to manually check identity clusters for specific users 2. Using Mixpanel's Data Pipelines to export your data to a warehouse where you can run more flexible queries 3. Contacting Mixpanel support for guidance on accessing identity cluster information programmaticallyIf you need a more precise solution, you might need additional information about how identity clusters are represented in Mixpanel's data model and accessible through JQL.

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