Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 

Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!

Reply
Anonymous
Not applicable

Google Analytics # users vs # users PowerBI

I am builing a dashboard to follow the GA numbers. What I see is that the numbers in GA are different than in PowerBI. The numbers in PowerBI are 10% higher. I pick the filtered data, the same as the view in GA.

 

Anybody knows how I could get the same numbers?

1 ACCEPTED SOLUTION
Anonymous
Not applicable

Here is an explaination AND a potential solution

https://www.skylinetechnologies.com/Blog/Skyline-Blog/October_2018/data-granularity-google-analytics...


The solution outlined makes everything a little more complicated (you'll have to create a DAILY fact, a WEEKLY fact, a MONTHLY fact etc.) *and* does not still resolve all possible use cases (if you want to report from Wednesday of week 1 to Tuesday of week 2, you still won't have the correct data)

View solution in original post

6 REPLIES 6
Anonymous
Not applicable

That is perfectly normal, you can't avoid it. Users are Unique Users.


The reason behind it is due to the way GA counts "unique users".

Imagine John Doe browsing your website at 23:30 of Day 1 and at 00:30 of Day 2. 

When you ask GA to count your users from day 1 to day 2, it will go into its database and search for UNIQUE users across those days. Looks like John Doe is the same person on day 1 and day 2, so it will report 1 Unique user.

However, PowerBI does not download the entire dataset of GA (it would be massive). What it does is querying GA for the data for every single dimension. So if you ask PBI to connect to GA and getting users From day 1 to Day 2 it will make two queries
- give me the users of day 1. GA answers "here you are, we have John Doe as unique on day 1"
- now give me the users of day 2. GA answers "Here you are, we have John Doe as unique on day 2"

And then PBI will sum users on day 1 + day 2 and will report 2 unique users, which is "wrong". Technically PBI will show the sum of unique users on the selected days, while GA will count exact unique users on day 1 and day 2.

This limitation CANNOT be avoided (*) and in fact means that any measure that is not summable on a dimension to be reported wrongly (for example Sessions and Users are wrong, while Page Views are correct).

This is the reason why, for our final customers, opted to use Google Data Studio instead of PowerBI to show GA data.

*= There is one way to avoid this, and it's to use a Direct Query connector to GA. Direct Query will re-query GA each time you make a new filter and GA will report correct number. But it has a cost AND when you switch to DQ everything gets a little more complicated.

Hi @Anonymous , how I connect to GA using Direct Query connector? Tks

Anonymous
Not applicable

(you can see this behaviour if you choose just one day in PowerBI, data will match GA)

Anonymous
Not applicable

Here is an explaination AND a potential solution

https://www.skylinetechnologies.com/Blog/Skyline-Blog/October_2018/data-granularity-google-analytics...


The solution outlined makes everything a little more complicated (you'll have to create a DAILY fact, a WEEKLY fact, a MONTHLY fact etc.) *and* does not still resolve all possible use cases (if you want to report from Wednesday of week 1 to Tuesday of week 2, you still won't have the correct data)

Anonymous
Not applicable

or you can spend 499/599$ /year

https://www.cdata.com/drivers/ganalytics/powerbi/
(note: cdata is quite well known, but i NEVER used this connector)

 

cdata doesnt help unfortunately...  it still doenst show the correct total users or sessions i na given time period. its the same as native connector which shows correct data by day but when ou want to see the overall users, its wrong

Helpful resources

Announcements
April AMA free

Microsoft Fabric AMA Livestream

Join us Tuesday, April 09, 9:00 – 10:00 AM PST for a live, expert-led Q&A session on all things Microsoft Fabric!

March Fabric Community Update

Fabric Community Update - March 2024

Find out what's new and trending in the Fabric Community.