Causes of Data Discrepancies Between Data Warehouse and Workspace

[MezzoMe]

Hello,

I noticed that for some time periods, the purchase numbers are different between Data Warehouse reports and Workspace dashboards in Adobe Analytics, even though they use the same data.

Could it be because of how the data is processed, sampling, or filters? I would really appreciate any advice on how to understand or fix these differences.

Thank you very much for your help.

[DanIMS1]

Hi @MezzoMe,

The data typically is the same in most cases but there can be some variations that can crop up.
Sampling is only really an issue in Adobe when you reach high volumes in dimension variations, and this only occurs in Workspace. Data Warehouse will use the full raw hit-level data, after it's been processed using processing rules and VISTA rules, nothing is sampled in Data Warehouse.

With the purchase event, in Workspace you can apply attribution settings to the metric. These may be different to the value in Data Warehouse. This should be checked in your case, just to rule it out.

If you are using segments in Workspace then these might not be applied in the data warehouse request, and there are some limitations in segments in Data Warehouse around sequential segments.

Depending on when you produce the request in Workspace and Data Warehouse there might be some variations in the data. For example if you have an offline source that adds data Adobe Analytics, you might request a Data Warehouse request one day and then look at the Workspace for the same date range later in the week after offline data was added after the Data Warehouse request made.
This would result in the historic information being updated in Workspace but not in the data from the Data Warehouse requests data. This can be checked up requesting another Data Warehouse request and comparing the data. If you are mixing App data with Web data this can happen a lot.

But as there can be multiple causes for differences, to accurately determine the exact cause for you we would need to know what exactly is being requested in Data Warehouse and how the report is generated in Workspace. 
It could be something as simple as the Data Warehouse is requesting data from a report suite whilst the Workspace is using a virtual report suite and then there is an additional filter that isn't being considered in the Data Warehouse request.

Hope this helps,
Thanks,

Dan

[MezzoMe]

Hi Dan,

Thank you so much for taking the time to provide such a thorough and detailed explanation — I really appreciate it.
As you mentioned, I have checked with the same conditions in both Data Warehouse and Workspace, but the same discrepancy still occurs.

Here’s what I have verified so far:

• Unified the extracted metrics and dimensions between Workspace and Data Warehouse (no segments used in either).

• Tested with data from a single day when the issue was first identified about a month ago.

• Found that for the same date, Workspace shows higher revenue than Data Warehouse (on the dates with issues, Workspace generally shows higher conversion performance).

• When checking by product performance, identified 1–2 missing conversions for specific products.

I’m not sure what other potential issues would be worth checking in this situation — could you please advise on any additional checks you would recommend?

Thanks again for your valuable input,

[harishkumar]

Ideally, it should not be different if you are pulling simple report without segment or multiple breakdown. you may need to ensure breakdown or segment used in workspace are in the same pattern as its in data warehouse. 

[Vinay_Chauhan]

Hi @MezzoMe 

In most cases, the data source is the same, but a few factors can cause mismatches -

• Attribution settings in Workspace can change metric values vs. raw Data Warehouse counts.

• Segments and filters may be applied differently, sequential segments in particular don’t always translate 1:1 to Data Warehouse.

• Timing of the request matters, Workspace always reflects the latest processed data, while a Data Warehouse file is a snapshot at the time it was pulled (so late-arriving/offline data could explain differences).

• Report Suite vs. Virtual Report Suite, check if Workspace uses a VRS with filters not applied in your Data Warehouse query.

If you align attribution, segments, date ranges, and the report suite selection, the numbers should be much closer.

Hope that helps!

[MezzoMe]

Thank you so much for your thorough and detailed explanation — I really appreciate your help.
As you mentioned, I’ve reviewed the data under the same conditions in both Data Warehouse and Workspace, but the discrepancy still remains.

Here’s what I’ve checked so far:

• Unified the extracted metrics and dimensions between Workspace and Data Warehouse (no segments used in either).

• Tested using data from a single day when the issue was first identified about a month ago.

• Found that for the same date, Workspace tends to show higher revenue than Data Warehouse (especially on dates where issues occur).

• When looking at product-level performance, I noticed 1–2 missing conversions for certain products.

I’m unsure if there are other angles I should explore — would you be able to suggest any additional checks?

Also, I’d appreciate it if you could elaborate a bit more on the Report Suite versus Virtual Report Suite concept, particularly how to verify if Workspace is using a Virtual Report Suite with filters that might not be applied in the Data Warehouse query.

Thanks again for your valuable guidance,