Creating a Visualization for Data from Two Days Ago
How can I make a visualization that contains only data from 2 days before today? Regardless of what today is it should still be data from two days ago.
Hi Mihir S., 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> đź‘‹
Feel free to ask a follow-up question once I've answered by mentioning me with @kapa.ai
To create a visualization in Mixpanel that contains only data from exactly 2 days before today, you can use the date range selector with the custom date options. Here's how to do it: 1. Open the report where you want to create your visualization (such as Insights) 2. Click on the date range selector (which typically shows a default like "30D" for last 30 days) 3. Select "Custom" from the dropdown menu 4. In the custom date selector, choose "Fixed" date range 5. Set both the start date and end date to be exactly 2 days ago from todayThis will create a fixed date range that includes only data from exactly 2 days ago. When you save this report to a board or as a bookmark, it will continue to show that specific date rather than dynamically updating.Alternatively, if you want a more dynamic approach where the date automatically shifts: 1. Select "Custom" from the date range dropdown 2. Choose "Last" option 3. Set it to "1" days 4. Click on "+ Ending" button 5. Set the ending to "2" days agoThis creates a rolling 1-day window that's offset by 2 days from today, effectively always showing "the day before yesterday" regardless of when you view the report.For more precise time control, you can also use the minute-level granularity option available in the advanced submenu under the custom date range selection, as mentioned in the Mixpanel documentation on date ranges.
kapa.ai is still learning and improving, please let me know how I did by reacting below.
ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.
Mihir S.··kapa.ai I have an event, I am looking at the frequency distribution per user. When I set the visualization to 30 days its giving me the total number of users that have 1,2,3,4,5,6,7 occurrences of the event but when i set the time period to 3 months or longer the distribution defaults to 1 and greater than 1 events. I can see the greater number breakdown anymore.
Let me check the knowledge base for you again. One moment! đź“š
I understand your issue with the frequency distribution visualization in Mixpanel. This behavior is related to how Mixpanel handles segment limits for different time ranges.When you extend your time range from 30 days to 3 months or longer, you're likely hitting Mixpanel's segment limits. Mixpanel implements these limits to maintain performance when analyzing large datasets. As mentioned in the documentation, Mixpanel truncates results to only return the top N segments when breakdown limits are reached.For Insights reports, there are specific segment limits: - Visible limit: 3000 segments - Selection limit: varies by visualization type (12-300 segments) - Default selection: typically 12 segmentsSegment Limits and Ordering documentation explains that these limits are in place "to keep the interactive analysis experience snappy for projects of all sizes."For frequency analysis specifically, you might be experiencing the limits described in the Frequency View documentation, where Mixpanel groups users into time-incremented buckets.To work around this limitation, you could: 1. Export your data as CSV, which has higher limits (up to 10,000 segments for most reports) 2. Apply filters to reduce the total number of segments 3. Break your analysis into smaller time chunksIf you need to see the detailed breakdown for longer time periods, the CSV export option is likely your best solution, as the documentation states: "For most reports, the CSV export limit is 10,000. For insights non-line charts, we allow for up-to 50,000."Would you like me to provide more specific guidance on how to export your frequency distribution data as CSV?
kapa.ai is still learning and improving, please let me know how I did by reacting below.
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! đź“š
I understand your frustration. Based on your description, you're seeing a very limited segmentation (just "1" and "greater than 1") even though your actual distribution segments are only 15 values (1-15), which is well below Mixpanel's segment limit of 3000.This behavior might be related to how Mixpanel handles behavioral properties and frequency distributions over different time periods. According to the documentation on behavioral properties: "Behavioral properties allow you to use your user's activity and use it as a property in your other analysis. Behavioral properties work like any other numeric property in that you are able to filter to a specific range of frequencies or customize how it is bucketed." Behavioral PropertiesFor longer time periods, Mixpanel might be automatically bucketing your frequency data differently to optimize performance, even though you're not hitting the absolute segment limits.Here are a few potential solutions: 1. Try using the "Frequency per User" option if available in your report type, which specifically shows "the number of times users performed an event" 2. Check if there's a bucketing or grouping setting in your report configuration that you can adjust to show more granular frequency data3. If you're using a Funnel report, the behavioral properties between steps might be affecting how frequencies are displayed4. Export your data as CSV to see if the full distribution is available in the exported dataIf none of these solutions work, this might be a specific limitation or behavior of the particular report type you're using. The documentation doesn't explicitly mention this specific truncation behavior for frequency distributions over longer time periods, so it might be worth reaching out to Mixpanel support for clarification on this specific behavior.Would you like me to provide more specific guidance on any of these potential solutions?
kapa.ai is still learning and improving, please let me know how I did by reacting below.
ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.