AdWords is now Google Ads. Our new name reflects the full range of advertising options we offer across Search, Display, YouTube, and more. Learn more

Data Studio
4.4K members online now
4.4K members online now
Ask questions about Data Studio
Guide Me
star_border
Reply

Discrepancies between Google Analytics and Datastudio concerning the number of users

Visitor ✭ ✭ ✭
# 1
Visitor ✭ ✭ ✭

All,

 

I am trying to set-up a Datastudio dashboard based on Google Analytics but I have a very significant discrepancy between Google Analytics and Datastudio concerning the number of users (30% more in Datastudio vs. Google Analytics) whereas the number of sessions or pageviews are correct.

Any idea why?

 

Thanks a lot for your help!

Discrepancies between Google Analytics and Datastudio concerning the number of users

Visitor ✭ ✭ ✭
# 2
Visitor ✭ ✭ ✭

Hi Romain: I have found that there is a delay in data population between GA and Data Studio. Make sure your data sources and associated views are the same between the 2. And use the refresh data tool in data studio to be sure.

Discrepancies between Google Analytics and Datastudio concerning the number of users

Visitor ✭ ✭ ✭
# 3
Visitor ✭ ✭ ✭

Thanks for your reply!

Indeed, it is now solved. What was surprising was that the Datastudio data were higher than Google Analytics and only for the metrics number of users...

Anyway, it now works like a charm!

Thanks again,

Romain

Discrepancies between Google Analytics and Datastudio concerning the number of users

Visitor ✭ ✭ ✭
# 4
Visitor ✭ ✭ ✭

Dear all,

In fact, it is not solved. It is working if I do a scorecard but it is not if I do a bar chart aggregated by week:

From March 20 to March 26 (week 12): 60 000 visitors based on the scorecard but 80 000 based on the bar chart.

Is there a potential issue with doubling counting returning visitors?

Thanks again for your help.

Romain

Discrepancies between Google Analytics and Datastudio concerning the number of users

Visitor ✭ ✭ ✭
# 5
Visitor ✭ ✭ ✭

Solved by using the GA field Iso Week (and not by recreating a field in DataStudio) !

Thanks again,

Romain

Discrepancies between Google Analytics and Datastudio concerning the number of users

Visitor ✭ ✭ ✭
# 6
Visitor ✭ ✭ ✭

I have the same issue with using "Quarter(Date)" as an created field to have the data shown by quarters of the year. Problem is also only concerning the number of visits - the others (sessions, time on site..) are right.

Discrepancies between Google Analytics and Datastudio concerning the number of users

Visitor ✭ ✭ ✭
# 7
Visitor ✭ ✭ ✭

Hey everyone - I have this same problem/question too. In talking with another friend, here's what we're thinking right now:

Basically, exported User data from Google Analytics (that's connected to Google Data Studio) can be different than what you see in Google Analytics (by a few users) because of how the data is compartmentalized in the export. To put it another way, it's because of how the data export is formatted. The same thing happens with a CSV export as the import to DataStudio.

The data that gets exported is was drives the graph in Data Studio, but those graphs usually have some options (hourly, day, week, month). The use of those options causes this disconnect.

Here's an example. Let's say I visit a website 12 times a year by visiting once per month. Each month, I'm counted as a unique user, but if you set the lookback window to 12 months then I only show up as one unique user. Could this cause the discrepancy we've been seeing with User data? 

This could be why Users change, but sessions and pageviews are the same. Sessions and pageviews are absolute. Unique users are relative to time, and the way Google Analytics exports data can't really deal with that completely.

Anyway, just think of Analytics as feeding data into Data Studio. That feed can handle absolutes, but it can't really handle things that change over time. Unique Users is something that changes over time based on reporting preferences, time settings, and graph settings.

 

Thoughts?

Discrepancies between Google Analytics and Datastudio concerning the number of users

Visitor ✭ ✭ ✭
# 8
Visitor ✭ ✭ ✭

I hear what you're saying @Matt L and that tends to ALWAYS be the issue when trying to rollup unsampled data over a longer period of time but I'm not sure that should apply for users in this case.  On paper, your scenario is correct where a User scoped metric spans over several days/sessions and if you separate those sessions in several data exports and then roll them up, you'll get an inflated total because the deduplication doesn't happen. And if that's how the data is exported into GDS then it would still hold true, however, GDS will simply kick into sampling if the data crunched becomes too large and then it'll properly process that smaller data chunk.

 

 I can see users off in GDS compared to GA when both claim no sampling is occurring. I have still yet to find a satisfactory answer to this despite checking for sampling, report caching, filters, data source, etc.  

Discrepancies between Google Analytics and Datastudio concerning the number of users

Visitor ✭ ✭ ✭
# 9
Visitor ✭ ✭ ✭

Hey Andy - yes, I totally agree that expecting numbers to tie-out completely is a fool's errand. But like you said, GDS to GA with the claim of no sampling is one of those weird spots. I'll let you know if I stumble across anything better. Thanks for digging in too!