We are getting 50% data discrepancy in Video events data from Mixpanel...

Mixpanel is headed to Money20/20 October 26 - 29 in Las Vegas, Nevada!

We are getting 50% data discrepancy in Video events data from Mixpanel... | Mixpanel Community

2 comments

· Sorted by Oldest

kapa.ai
APP
·
·
Hi Mansi g., 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
·
·
A 50% data discrepancy between Mixpanel and Brightcove for video play events (500 in Mixpanel vs. 1000 in Brightcove) can be caused by several factors. Based on the knowledge sources, here are the most likely reasons: 1. Different Event Definitions and Triggers The two systems may define or trigger a "video play" event differently. For example, Mixpanel might track the event when the SDK loads or when a user clicks play, while Brightcove could trigger it at a different point (e.g., when the video actually starts streaming). If the triggers are not aligned, the counts will differ. It's important to ensure both systems are tracking the same user action at the same point in the user journey Debugging Discrepancies.2. Client-Side vs. Server-Side Tracking and Ad Blockers If Mixpanel is tracking events client-side (in the browser or app), ad blockers or privacy settings (like Do Not Track) can prevent events from being sent, leading to undercounting—sometimes losing events for 30-50% of users. Brightcove may use server-side tracking, which is less affected by these issues Data Discrepancies.3. Different Timeframes, Filters, or Properties Discrepancies can also arise if the reports are using different timeframes, filters, or are counting different properties (e.g., total events vs. unique users). Double-check that both systems are reporting on the same time period, with the same filters, and are counting the same metric Data Discrepancies.4. Delayed or Missed Event Ingestion Mixpanel may not ingest events that are sent more than 5 days after they occurred, while other systems might accept late events. Also, network issues or implementation bugs can cause events to be missed or delayed in Mixpanel Data Discrepancies.5. Duplicate or Skipped Events Technical issues in event tracking code can cause events to be double-counted or skipped in either system. For example, Mixpanel might miss events if the tracking code is not triggered correctly, or Brightcove might count duplicate plays if the event is fired multiple times for a single user action Product analytics implementation hiccups that are easy to spot.What to do next: - Align the event definitions and triggers in both systems. - Check for client-side tracking issues (ad blockers, privacy settings).
- Ensure both systems are reporting on the same timeframe and filters. - Investigate the implementation for duplicate or missed events. - Consider implementing server-side tracking for more reliable data.If after these checks the discrepancy persists, drill down into specific user journeys or segments to identify where the counts diverge 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.

kapa.ai
APP
·
·
Hi Mansi g., 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
·
·
A 50% data discrepancy between Mixpanel and Brightcove for video play events (500 in Mixpanel vs. 1000 in Brightcove) can be caused by several factors. Based on the knowledge sources, here are the most likely reasons: 1. Different Event Definitions and Triggers The two systems may define or trigger a "video play" event differently. For example, Mixpanel might track the event when the SDK loads or when a user clicks play, while Brightcove could trigger it at a different point (e.g., when the video actually starts streaming). If the triggers are not aligned, the counts will differ. It's important to ensure both systems are tracking the same user action at the same point in the user journey Debugging Discrepancies.2. Client-Side vs. Server-Side Tracking and Ad Blockers If Mixpanel is tracking events client-side (in the browser or app), ad blockers or privacy settings (like Do Not Track) can prevent events from being sent, leading to undercounting—sometimes losing events for 30-50% of users. Brightcove may use server-side tracking, which is less affected by these issues Data Discrepancies.3. Different Timeframes, Filters, or Properties Discrepancies can also arise if the reports are using different timeframes, filters, or are counting different properties (e.g., total events vs. unique users). Double-check that both systems are reporting on the same time period, with the same filters, and are counting the same metric Data Discrepancies.4. Delayed or Missed Event Ingestion Mixpanel may not ingest events that are sent more than 5 days after they occurred, while other systems might accept late events. Also, network issues or implementation bugs can cause events to be missed or delayed in Mixpanel Data Discrepancies.5. Duplicate or Skipped Events Technical issues in event tracking code can cause events to be double-counted or skipped in either system. For example, Mixpanel might miss events if the tracking code is not triggered correctly, or Brightcove might count duplicate plays if the event is fired multiple times for a single user action Product analytics implementation hiccups that are easy to spot.What to do next: - Align the event definitions and triggers in both systems. - Check for client-side tracking issues (ad blockers, privacy settings).
- Ensure both systems are reporting on the same timeframe and filters. - Investigate the implementation for duplicate or missed events. - Consider implementing server-side tracking for more reliable data.If after these checks the discrepancy persists, drill down into specific user journeys or segments to identify where the counts diverge 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.