Data analysis is like a puzzle or a mosaic. A single piece provides almost no useful information, but properly assembled, multiple data sources can create a complex and insightful picture. Similarly, visualising how multiple data points from various sources interact and evolve over time can help answer important questions about user behaviour and allow us to see important data patterns.

At the beginning 

From the start of the project, the main data source available was our Europa analytics tool , and the main focus for reporting was on profile and behaviour of users.

Monitoring the various data points offered by the tool over time (visits, page views, countries of origin, most visited content, referrers etc) allowed us to identify which pages and websites were attracting traffic, which were redundant and also get a few insights into who our visitors were.

We then began to combine different data sources in order to get a better view of our audience. Initially, social media and search data were integrated into our reporting. The data was handled using R (a software environment for statistical computing and graphics), and gave us good insights into how search, social and the websites interacted.

The team went on to launch one of the largest on-line polls in Commission history. Those findings focused on the main reasons why people interact with the European Commission's website. Beyond this however, a wealth of additional information was also collected (preferred language, country of origin, section they were visiting, profession etc). This helped enrich our understanding of our audience, and, for example, gave us the data necessary to prioritise languages for translation. 

Where we are now

As we prepare the launch of the Beta version of the Commission's new web presence, we are making sure we have all the tools in place to measure its performance. We are also trialling a new open source analytics system for the beta -  Piwik.

Our goal is to have a constant and systematic integration of multiple data sources to help us report on and measure the effectiveness and impact of our digital communications. We will observe the patterns, determine how the various platforms interact, establish correlations, assess what users want to do, all in order to better serve their needs.

Further reading:

https://hbr.org/2014/05/10-kinds-of-stories-to-tell-with-data

https://www.thinkwithgoogle.com/articles/measure-what-matters-most.html

https://hbr.org/2013/03/advertising-analytics-20

 

Related disciplines: 


Add new comment