In this example, we assume you setup Google Analytics as a data source within the Stitch UI. One of the first steps in that process is to name the integration.
In the screen below Stitch will ask you to name the integration:
You will want to use a naming convention for any integration you setup. In our example ( 1 ) we show a template of
ga_<property identifier>_<report identifier> . Using consistent naming conventions goes a long way in making sure your warehouse is well organized.
- Since we are setting up Google Analytics, we will want to use
gaas the prefix
- Next, we want to make sure the property name is set. If my property name was called Openbridge, then we should use that. This will give is
- Last, we will want to set the report name. This reflects the specific data being collected as part of the integration. For example, if you wanted to collect event data, you would have an integration name of
ga_openbridge_events. If you want to collect Channel and Campaign data, name the integration
You are free to choose a naming convention that works best for your situation. Always remember, naming your integration something unique is important since it will be used as the basis for table and view names in your warehouse.
Let's assume you called your integration within Stitch
ga_openbridge_events . This name aligns with the data you are collecting.
With Amazon S3 as your destination, Stitch will use the destination template you setup:
This will result in delivery of data to your account
youruniqueID001, with an integration name called
ga_openbridge_events . Stitch will output all the CSV data according to your object key template:
As we stated previously, we use the integration name to define the table name. This means CSV data will be loaded into a table called
You would repeat this process for every subsequent integration you want to set. For example, if you had setup another integration called
ga_openbridge_channel_campaign the data would be delivered as follows:
Following our naming conventions, the data will be loaded into a table called