Forum Discussion
google analytics
Hello!
First off, thanks for reading and sharing your expertise. I'm a new Power BI user and running into some trouble with Google Analytics.
I connected GA to the desktop version of Power BI pulling in:
- % Exit
- Avg. time on Page
- Date
- Default Channel Grouping
- Page
- Page Title
- Unique Pageviews
I did a quick test comparing unique page views for the month of September. The data in BI doesn't match GA. What could possibly be causing this error?
The other issue I'm running into is with the date hierarchy. The date from GA includes the day, month, date, and year. When I add the date column as a filter and select date hierarchy, my only option is year. Why is month not also included? Is there a work around to add this an option?
Hi carlydugger,
The connector in Power BI Desktop rely on the Google Analytics Core Reporting API. Although we attempt to release updates in accordance with any changes, the API may change in a way that affects the results of the queries we generate. Due to this dependency we cannot guarantee the results of your queries when using this connector.
In your scenario, please try to update the Power BI Desktop to the latest version 2.39.4526.362 and connect to Google Analytics. If issue persists, according to this article, this should be a better place for you mainly discussing questions related to google-analytics-api.
In addition, we are not able to place a date hierarchy in Visual /Page /Report level filter currently. To work around the issue, you can use the custom visual HierarchySlicer.
Best Regards,
Qiuyun Yu
7 Replies
I ran into the same problems.
Connecting to the GA API via Power BI gave me wrong numbers while using the GA API via SEO Tools for Excel was ok.
This may be linked to the amount of data imported.
I have tried to modify the query using the advanced editor.
I added a line with samplingLevel= "LARGE", as indicated in the documentation
let Source = GoogleAnalytics.Accounts(), Account = Source{[Id=#"GA Account Id"]}[Data], Property = Account{[Id=#"GA Property Id"]}[Data], View = Property{[Id=#"GA View Id"]}[Data], samplingLevel= "LARGE", #"Éléments ajoutés" = Cube.Transform(View, {{Cube.AddAndExpandDimensionColumn, "ga:yearMonth", {"ga:yearMonth"}, {"Month of Year"}}, {Cube.AddAndExpandDimensionColumn, "ga:medium", {"ga:medium"}, {"Medium"}}, {Cube.AddMeasureColumn, "Sessions", "ga:sessions"}}) in #"Éléments ajoutés"It does not return errors, but seems to solve the issues as the figures are the same as in GA.
However, another solution would be to trim the imported data to a selection of dates.
With the GA API, you have to set a start and end date
In Power BI, you always seem to import everything.
I have tried to add dateranges to my query as per this guide
dateRanges= {[startDate="2014-11-01", endDate="2014-11-30"]},It does not throw any error, but the date ranges are not applied either.
If this is possible, what is the right syntax?
Hi carlydugger,
The connector in Power BI Desktop rely on the Google Analytics Core Reporting API. Although we attempt to release updates in accordance with any changes, the API may change in a way that affects the results of the queries we generate. Due to this dependency we cannot guarantee the results of your queries when using this connector.
In your scenario, please try to update the Power BI Desktop to the latest version 2.39.4526.362 and connect to Google Analytics. If issue persists, according to this article, this should be a better place for you mainly discussing questions related to google-analytics-api.
In addition, we are not able to place a date hierarchy in Visual /Page /Report level filter currently. To work around the issue, you can use the custom visual HierarchySlicer.
Best Regards,
Qiuyun YuThank you! The custom slicer is wonderful. I'll continue to search for API solution elsewhere.
I ran into the same problems.
Connecting to the GA API via Power BI gave me wrong numbers while using the GA API via SEO Tools for Excel was ok.
This may be linked to the amount of data imported.
I have tried to modify the query using the advanced editor.
I added a line with samplingLevel= "LARGE", as indicated in the documentation
let Source = GoogleAnalytics.Accounts(), Account = Source{[Id=#"GA Account Id"]}[Data], Property = Account{[Id=#"GA Property Id"]}[Data], View = Property{[Id=#"GA View Id"]}[Data], samplingLevel= "LARGE", #"Éléments ajoutés" = Cube.Transform(View, {{Cube.AddAndExpandDimensionColumn, "ga:yearMonth", {"ga:yearMonth"}, {"Month of Year"}}, {Cube.AddAndExpandDimensionColumn, "ga:medium", {"ga:medium"}, {"Medium"}}, {Cube.AddMeasureColumn, "Sessions", "ga:sessions"}}) in #"Éléments ajoutés"It does not return errors, but seems to solve the issues as the figures are the same as in GA.
However, another solution would be to trim the imported data to a selection of dates.
With the GA API, you have to set a start and end date
In Power BI, you always seem to import everything.
I have tried to add dateranges to my query as per this guide
dateRanges= {[startDate="2014-11-01", endDate="2014-11-30"]},It does not throw any error, but the date ranges are not applied either.
If this is possible, what is the right syntax?
Hi offrench,
I imagine is because you have to add the samplingLevel on the cube request, not outside. As you can see in the documentation that you posted.POST https://analyticsreporting.googleapis.com/v4/reports:batchGet { "reportRequests": [ { "viewId": "XXXX", "dimensions": [{"name": "ga:medium"}], "metrics": [{"expression": "ga:sessions"}], "samplingLevel": "LARGE" } ] }
Still, somehow the cube request still not work because it doesn't have the same request, maybe you will have to do some work around.
Best,Jorge
carlydugger, I have the same problem with the data of GA, It´s something related with the API of GA and the quantity of data.
The workaround for that is, filter by dates periods in Power Query until your data match in PBI and GA.And we will wait to Power BI team fix this problem
