A look at the ideas and projects we're working on as we strive to grow and improve ReliefWeb.

ReliefWeb Data Snapshots: RW usage analysis powered by ReliefWeb API and Google Analytics

By Miguel Hernandez Rivera

RW Data Snapshots header

ReliefWeb received, on average, about 1.3 million visits a month in 2016. Our users visit the site to search for the latest crisis information, job announcements and opportunities to brush up their skills and knowledge. They access the site from the field and headquarters, using various devices.

As an operator of a major humanitarian information platform, we have been improving our Google Analytics data gathering in order to improve our services based on user behaviors. The continuous improvements to our API in the past years have made it easier to analyze what’s in our content repository with information dating back to 1996, which we made available to users via Content Trends alpha in 2014.

RW Data Snapshots screenshot

Now, we have combined the power of these two data sources, so that you, as a reader or a content provider, can analyze ReliefWeb usage through ReliefWeb Data Snapshots alpha. With this new tool, you can see how many people access ReliefWeb content online, which countries, themes or disasters are the most read, and where users are based, their main language, their device and browser, and user attributes. Because it uses Google Analytics, individual visitors are not tracked across websites and visitors' IP addresses are anonymized before they are ever stored.

The application has two distinct focuses: amount of Content Published vs. Number of Sessions.

You can choose between seeing the amount of content published or the number of sessions on ReliefWeb. ReliefWeb performs systematic tagging of each content with a set of “categories”. As a result, with the [# of] content published operator, you can get some information and trends about the disasters with the highest number of reports, themes that have trended between certain dates, or the job categories most sought after for certain duty stations. If you choose the [# of] sessions operator for the main filter, you will be able to see the readership of different content by ReliefWeb categories, which are governed by our taxonomy.

RW Data Snapshots work well on mobile devices, tablets and desktop screens and you can download any of the datasets in CSV format.

And now for some technical stuff… as mentioned earlier, the data comes from ReliefWeb Google Analytics account through its API (for counting the visits) as well as ReliefWeb API (for counting the amount of content). The categories exposed have been tracked using Google Analytics' custom dimensions, dating back to April 2016. The application is hosted on GitHub public pages and we have a little proxy and cache system to speed up Google Analytics queries hosted in a cloud provider.

If you want to know more about the RW Data Snapshots, please visit the application wiki where you will find some interesting user cases to inspire you. We would be very interested in hearing about your experience using ReliefWeb’s content and analytics data - please share them in the comments!