Should We Exclude Outlier Values in Our Presentation Data?

Rachel V. · 2025-09-10T21:01:55.469Z

I’m not sure it makes sense to run all the detailed checks right now. For the presentation, would it be acceptable to exclude the outlier value, or does this indicate that the chart itself may not be reliable? What do you think?

Should We Exclude Outlier Values in Our Presentation Data? | Mixpanel Community

2 comments

· Sorted by Oldest

kapa.ai
APP
·
·
Hi Rachel V., 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
·
·
Short answer: yes, you can exclude clear outliers for a presentation—but do so transparently, and make sure the underlying trend is still representative.- If precision isn’t critical for this metric (e.g., page views, clicks), de-emphasizing outliers can be acceptable to communicate the broader trend. Mixpanel even supports speed/accuracy tradeoffs (e.g., query-time sampling) for exploratory views. [Warehouse journeys.] - However, a big outlier can be a symptom of data quality issues (duplicates, sequencing, misconfiguration). If that’s the case, the chart may not be reliable until you sanity-check key events. [Business impacts.] - Quick checks you can do now: - Look for duplicate/spiky event bursts in the same second (a common bug). [Technical hiccups.] - See if “undefined/null” segments are inflating results; exclude them in the legend or filter “is set.” [Falsy values.] - Use Insights percentiles (p90/p99) instead of averages to reduce skew in metrics like response or cart values. [Aggregation examples; Zendesk in Mixpanel.]Recommendation: - For the presentation: show the primary chart with outliers excluded or with a percentile-based view, and add a small note on the exclusion criteria. Then include a backup slide showing the raw view for transparency. - Afterward: prioritize a light audit on the affected event(s) to confirm it’s not a systemic issue before relying on the chart for decisions. [Business impacts—prioritize.]
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.

kapa.ai
APP
·
·
Hi Rachel V., 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
·
·
Short answer: yes, you can exclude clear outliers for a presentation—but do so transparently, and make sure the underlying trend is still representative.- If precision isn’t critical for this metric (e.g., page views, clicks), de-emphasizing outliers can be acceptable to communicate the broader trend. Mixpanel even supports speed/accuracy tradeoffs (e.g., query-time sampling) for exploratory views. [Warehouse journeys.] - However, a big outlier can be a symptom of data quality issues (duplicates, sequencing, misconfiguration). If that’s the case, the chart may not be reliable until you sanity-check key events. [Business impacts.] - Quick checks you can do now: - Look for duplicate/spiky event bursts in the same second (a common bug). [Technical hiccups.] - See if “undefined/null” segments are inflating results; exclude them in the legend or filter “is set.” [Falsy values.] - Use Insights percentiles (p90/p99) instead of averages to reduce skew in metrics like response or cart values. [Aggregation examples; Zendesk in Mixpanel.]Recommendation: - For the presentation: show the primary chart with outliers excluded or with a percentile-based view, and add a small note on the exclusion criteria. Then include a backup slide showing the raw view for transparency. - Afterward: prioritize a light audit on the affected event(s) to confirm it’s not a systemic issue before relying on the chart for decisions. [Business impacts—prioritize.]
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.