Understanding Filter Discrepancies in Metrics Analysis | Mixpanel Community

Understanding Filter Discrepancies in Metrics Analysis

I am getting different results when I'm using filters in the filter section as opposed to using filters in the Metrics Section. I already have a first time ever filter selected in the Metrics Section. Why the discrepancy?

3 comments

kapa.ai
APP
·
·
Hi Aneem I., 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> 👋
Feel free to ask a follow-up question once I've answered by mentioning me with @kapa.ai
kapa.ai
APP
·
·
The discrepancy you’re seeing between using filters in the filter section (global filters) and filters in the Metrics section (inline or event-level filters) is due to how and when these filters are applied in Mixpanel’s query logic.Key points:Why the discrepancy? If you have a "First Time Ever" filter in the Metrics section and then apply an additional filter in the global filter section, the order and scope of filtering can lead to different results. For example, the "First Time Ever" filter may select a different set of events than what the global filter would include or exclude after the metric is calculated. This is especially true if your filters are based on properties that may or may not be present on the first event for each user.Best Practice: To avoid confusion, try to keep your filtering logic consistent—preferably apply all relevant filters at the metric/event level when working with "First Time Ever" or similar filters, so you’re always working with the intended event set.For more details, see: - Analyzing First Time Users - Filters in Reports - Query Builder FeaturesIf you need to debug further, try removing all filters and re-adding them one by one to see where the discrepancy arises Debugging Discrepancies.
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.
Aneem I.
·
·
Best Practice: To avoid confusion, try to keep your filtering logic consistent—preferably apply all relevant filters at the metric/event level when working with "First Time Ever" or similar filters, so you’re always working with the intended event set. Why should I do this?