How to handle Deduplication in Metric Aggregation Over Multiple Dates (we fetch data from Adobe Analytics via API for individual dates) | Community
Skip to main content
J
JESBINSO1
April 16, 2024
Solved

How to handle Deduplication in Metric Aggregation Over Multiple Dates (we fetch data from Adobe Analytics via API for individual dates)

  • April 16, 2024
  • 1 reply
  • 947 views

Hi,

We’re currently retrieving metric values for specific dimensions through the Adobe Analytics API for individual dates. However, we’ve encountered a discrepancy when aggregating these metrics over a date range. Our calculated aggregate doesn’t align with the one reported by Adobe Analytics, likely due to the deduplication process within the tool. Could someone provide guidance on reconciling this aggregation inconsistency between our calculations and those of the Adobe Analytics platform?

 

Thanks in Advance,

Jesbin

This post is no longer active and is closed to new replies. Need help? Start a new post to ask your question.
Best answer by Nick_Walter

As far as I know, unless you can pull data at the user level where you can also pull in a unique ID for each user and use that to dedupe, there isn't a good solve for this. It is hard to pull data at this level of granularity using the API because of the row limitation.

My first question is, if you need this level of data could you build whatever report this is in Adobe workspace rather than in another tool?

Second, is it important to have these numbers be the same when you know what the discrepancy is. This is a case where either report is wrong, one just may be "more correct" depending on the business question you are trying to solve for.

Finally, are you using CJA or AA? If CJA, they are making a new connector that will eliminate the 50k row restriction.

1 reply

Nick_Walter
Community Advisor and Adobe Champion
Nick_WalterCommunity Advisor and Adobe ChampionAccepted solution
Community Advisor and Adobe Champion
April 16, 2024

As far as I know, unless you can pull data at the user level where you can also pull in a unique ID for each user and use that to dedupe, there isn't a good solve for this. It is hard to pull data at this level of granularity using the API because of the row limitation.

My first question is, if you need this level of data could you build whatever report this is in Adobe workspace rather than in another tool?

Second, is it important to have these numbers be the same when you know what the discrepancy is. This is a case where either report is wrong, one just may be "more correct" depending on the business question you are trying to solve for.

Finally, are you using CJA or AA? If CJA, they are making a new connector that will eliminate the 50k row restriction.

    Terms & ConditionsAccessibility statement

    Loading footer...