ChannelMix Dataset QC & Discrepancy Troubleshooting

What do I do if my data looks incorrect?

Check out the steps below to help identify any possible issues.

Step 1: Make sure you’re viewing the dataset at the correct level of granularity/dimensionality.

  • This means that when you’re comparing a ChannelMix dataset to a user interface (UI) reporting dashboard, make sure both have the same dimensions such as campaign, adgroup, keyword, ad, geo, device, and other dimensions that might have an impact on how data will be returned.
    • A lot of times, a user will be looking at an adgroup level dataset from ChannelMix and comparing it to the ad level report in the UI. In reality, they need to be checking the adgroup level report in the UI.
  • Also, make sure the UI data is at the daily level and not aggregated to monthly or some other span.
  • Note: If you need help extracting sample reports for Google Analytics, try the Google Analytics Query Explorer.
  • What can I do?
    • You can activate a standard dataset that meets the level of granularity or submit a ticket through ChannelMix Control Center for a custom dataset request.

Step 2: The issue could be related to how recently data has been pulled from the source system

  • ChannelMix datasets do not contain live data and are usually updated in the morning, so if you’re comparing ChannelMix data to live UI data in the afternoon, there could be some slight differences.
  • What can I do?
    • You can Initiate a dataset refresh through the ChannelMix Control Center Library page to re-pull data or wait until the next day for the data to be refreshed automatically.

Step 3: The issue could be related to attribution windows

  • Each data source has a different attribution/conversion window. You will need to compare our dataset with the same attribution window:
    • For example, the default for Facebook is 1-day view and 7-day click. But if you’re viewing your performance in your Facebook Ads Manager on a 7-day view and 7-day click, then your data will not match.
  • You need to compare the dataset past the attribution window because the live data in UI will keep on changing until it's outside of the window.
    • For example, if your attribution window is set to 7-day view and 7-day click, you will need to compare the data past 7 days since the conversion window has not closed yet.
  • What can I do?
    • If you do not utilize the default attribution window for a source system, let us know and we can make adjustments to the data pulls, so that they align with your reporting strategy.

Step 4: What's Next?

  • If the steps above don't help uncover the root of the problem, send a ticket our way and provide us with as much information as you can, including any data exports that were referred to above, and we'll take a closer look.
Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request



Please sign in to leave a comment.