On December 6 and 7, 2016, Google invited us to an in-depth workshop on their big data technology: BigQuery. This workshop took place over the course of two mornings in London.
We already use BigQuery at altima° with two primary perspectives. The first is to use the Analytics connector supplied by Google to complete the Big Query data sets with Analytics data in order to process them without the classic interface.
We also use BigQuery to stock the results of API requests, notably Google’s. The object of these requests is to recover relevant data from different sources that will then be aggregated.
Most of the time, we use the data present in BigQuery in Data Studio to create complete and personalized dashboard reports. We have also developed a system that emails reports monthly.
We currently make requests to BigQuery by way of scripts written in node.js.
This workshop allowed us to discover new ways to use BigQuery specifically by requesting e-commerce data directly using a language that largely resembles SQL, and permits excellent performance results with a single request.
On another level, during this workshop we used The Jupyter Notebook which makes it possible to combine executable code, mathematical functions and markdown within the browser.