How to visualise Windsor.ai (Google Ads, Facebook, Bing, Google Analytics, DoubleClick, …) data in Datorama

Post Reply
niklaskolster
Posts: 26
Joined: Mon May 18, 2020 7:33 pm

How to visualise Windsor.ai (Google Ads, Facebook, Bing, Google Analytics, DoubleClick, …) data in Datorama

Post by niklaskolster »

The Windsor.ai platform supports API calls to retrieve and send information. In this article, we explain how to extract data and visualise it in Datorama.

The prerequisites are:
  • Windsor.ai account
  • Windsor.ai API key
  • Datorama Account with Administrator privileges
Here are the steps required to import data into Datorama
  • In Datorama head to Connect & Mix -> Data Streams and create a new Data Stream. Select HTTP under Technical Vendors
    Image
  •  Fill out the following details
    Name: Windsor.ai - Data Stream
    Data Stream Type: Web Analytics
    Default Channel Name: Attribution Data
    Image

    If Attribution Data does not exist Click on the Add new Button next to Default Channel Name and add it
    Image
  • Build the connection string and fill it into the connection info section
    Image

    Here is an example query:

    Code: Select all

    https://api.windsor.ai/pied_piper_llc/pied_piper_llc_attribution/public/pied_piper_llc_attributions_and_costs?api_key=abcd
    &_select=date,source,medium,campaign,totalcost,type,markov,markov_value
    &_groupby=date,source,medium,campaign,totalcost,type,markov,markov_value
    &_order=-date
    The highlighted fields:
    – pied_piper_llc should be replaced with your username
    – pied_piper_llc_attributions_and_costs should be replaced with the table you want to load
    – abcd should be replaced with your API key
    – &_select=date,source,medium,campaign,sum:totalcost should be replaced with the columns and metrics which you want to visualise
    – &_groupby=date,source,medium,campaign should be replaced by the columns you want to group by
    – &_order=-date should be replaced by the columns you want to order by

    The API documentation can be found here and the available fields can be found in the table definitions  here
  • Click on Validate to make sure that the query is working
    Image
  • Leave the Frequency and Deliver Hour as it is and click on Next
    Image
  • Create the mappings (this step will depend on your data structure).

    Image

    Important: The date needs to be parsed so it fits the date in Datorama. To do so you will need to apply this formula to the date column 

    Code: Select all

    PARSEDATE(csv['date'],'yyyy-MM-dd')
    Image
  • Click on Save. You will now get a Data preview

    Image
  • Your Windsor.ai data should now appear

    Image
  • You can now start visualising the data in your Datorama dashboards!
    If you would like to build another query to get different dimensions and metrics into Datorama please have a look at our API documentation, which can be found here, and our table definitions, which can be found here.
Post Reply