Forum Discussion
PowerBi online limits using Google Analytics?
Thanks for your reply smoupre.
I mean if I use, for example, three years (Data range) 50,000 rows of data appear to me. If I create a data table with powerbi, ¿will it be able to reflect me these 50,000 rows? It has a maximum data to display the table that I created? If the maximum are 10,000 rows (for example) in the table you could not see the 40,000 rows missing. To use the query of Google Analytics I go to: https://ga-dev-tools.appspot.com/query-explorer/ .
And with respect to the first question, I like to use the data that interests me, no predefined dashboard.
The number of rows the underlying PowerPivot engine can handle is quite high. I have seen files with 100mio+ rows. It gets quite slow depending on the calculations of your measures but its possible.
If you query a lot from Google Analytics it might take some time to load the data and you should watch your query limits but with updates once a day you are fine.
Not sure if I understand your questions correctly.
Why would you display a table with more than 10k rows? Is there a real necessity for this / would an end user want to scroll down thousand of rows?
Most people would be better of with aggregated tables, lets say "sessions by source site" which can then be filtered using slicers.
Or do you just want to handle the 50k rows in an aggregated visual? Depending on the visual there are some upper limits. Funnel displays for example can display dozens of values but become unreadable due to the shrinking bar sizes.. some regular tables without aggregation features can show a few hundred values but seem to cut of rows after that. The calculations are not affected by that - just the visual itself.
I found that by connecting directly to Google Analytics from the Power BI service works well for small datasets. It provides a nice set of reports and dashboards that are mostly helpful. However on larger datasets, the refresh times out and there is no way to change the timeout settings.
By creating a Power BI Desktop report and then connecting you get more control over what measures and dimensions you actually want to use. You can also combine these at this point with other data sources. eg. We only bring the Date value from GA and then combine with a local Date Dimension table to get reporting periods that suit our business and not google.
I've yet to have a timeout issue when using Power BI Desktops connector to GA and this is my preferred method.
