QUICK ANSWERS TO COMPLEX QUESTIONS
At Improving Metrics we always talk about the importance of data, how to make a good collection and how to turn it into useful knowledge for companies, but what happens when we handle volumes of inordinate data? Have you ever wondered how we manage to organize it? Or how can we query it easily?
At first, this may seem somewhat complex, and closer to Big Data than Digital Analytics, but it is something that happens every day in any company specializing in digital analytics.
What differentiates a good analysis of data is the ability to relate all this data, coming from the different sources, with the analysis of the traffic our users generate on the site. That's why, once again, Google offers us an easy solution with Google BigQuery, a tool to give quick answers to complex questions.
This tool allows any company to use Google's own processing capability, with its own Data Processing Centers (CPDs) around the world, so to balance capabilities and provide them with a superior speed access regardless of where the queries are made.
BigQuery offers you an already configured Big Data cloud system, so you only have to worry about sending the data from your systems to BigQuery and it will store them for when you need to query it. This way, when you want to perform an analysis, the tool provides you with the facility to make any query and get results in seconds, regardless of the volume of your data.
BIGQUERY STRUCTURE AND HOW IT WORKS
The complexity of the questions you ask can go as far as your imagination is able to reach, as you can give answers to topics whose information you do not have available in the first instance.
For example, you may wonder not only if your sportswear-selling eCommerce sells more during weekdays or weekends, but also know if there are more purchases when it is warm or cold, or if sales on weekends increase when there are sports events going on.
BigQuery uses SQL queries, very fast type of queries that run thousands of megabytes of data in seconds. In addition, these queries are cached for 24 hours, so if you re-launch that query, the data will return immediately.
The great potential of BigQuery is that you will forget about the server infrastructure. Google will do everything so you can relax. They care about hosting the information and optimizing it so that you only have to worry about analysing it.
Let's suppose we want to know the number of queries that have been made in Wikipedia about the films nominated for the Oscars, during the week of the awards. Access to Wikipedia data is public and can be easily integrated into BigQuery, well, that "simple" query will process 300GB of information, and with Big Query, you get the result in just 7 seconds.
Imagine if you had to set the necessary environments up to be able to process those 300GB at that time, we would be talking about many servers to be able to do it.
SIGN-UP AND INTEGRATION WITH OTHER TOOLS
Google makes available to each new user a series of initial queries to test the full potential of the tool.
For Google Analytics Premium or Google 360 users, there is a 500 € of "give away" queries per month, and from there, you'll only have to pay a reduced price for storing all your information.
The pricing model can be found here in more detail.
As for the sign-up and integration with the vast majority of Google tools (Analytics 360, Adwords, DFP, Doubleclick ...) cannot be simpler. You can register and start operating instantly after just 4 clicks. Just select the option to bring the data to BigQuery from any of the other tools and that's it.
In addition, you can update the data without limitations and most of Google's tools are integrated directly so that the information is dumped in real-time.
BIGQUERY AND GOOGLE ANALYTICS 360
Integrating BigQuery with Google Analytics 360 allows you to extract raw information from Google Analytics and make much more complex queries without the limitations of tables or "pre-made" reports in Google Analytics.
The great advantage is that you can cross Google Analytics information without the limitations of tables and reports and enrich that data with the sources of information you want in order to extract much more useful information.
For example, a pizza sales company used BigQuery to detect peaks of sales at each of their physical stores.
With this data, they developed a predictive model for Google Analytics to create segments with a custom dimension and communicate with Google Doubleclick Bid Manager to optimize investments in advertising.
So, they launched the campaigns in the moments that they knew there would be more demand based on their predictive analysis, and they managed to reduce advertising revenue by 40% while boosting their sales.Right ads at the right moment.