Blend not providing same amount of data as direct connection

[Toni03]

I am testing the difference between data received by using blends vs using a direct connector. I have a direct connection to a Google BigQuery table. And I also have a blend to that same connection to the BigQuery table. There is just that connection in the blend, not another table, no join, no anything.

However, my direct connection returns 12,780 rows, while The blend returns only 4,725 rows. The configuration is the same. I can also see which data is missing and I can not find any pattern. The missing data is not only from the end or just the beginning of the query, it's dispersed.

Does anyone have any ideea on why this could happen?
Does the blend maybe reduce the result for any reason?
Could it be maybe because I am not supposed to use a blend if I have only one table in it (so not joining any 2 or multiple tables)?

Thank you

[Kirbs]

Hi there,

Is it possible that you are unintentionally hiding fields that would then decrease the amount of data in your blend?  You can check this in your blend configuration. 

Otherwise, we have general tips on data blending that can be found here  and troubleshooting steps for blends that can be found here, though most of these steps apply to blends with more than one data source. 

[Toni03]

Thank you @Kirbs for your response!
Repeated join fields do not exist, since there is only one table, but I unchecked it anyways just to be sure. 
I my debugging I tried to create two identical belnds, both being connected to the same table, and having the same Dimensions, and Metrics. So basically 2 identical blends. However the results are not the same.

I have also observed that the issues appear when I start adding the GEO type dimension, which in my case is a country code. If I don't have the country column as a dimension there is no issue