{"id":367,"date":"2017-09-29T10:42:21","date_gmt":"2017-09-29T10:42:21","guid":{"rendered":"https:\/\/www.ebi.ac.uk\/about\/clusters\/technical-services\/?p=367"},"modified":"2022-06-25T20:21:46","modified_gmt":"2022-06-25T20:21:46","slug":"web-analytics-how-we-collect-data","status":"publish","type":"post","link":"https:\/\/www.ebi.ac.uk\/about\/teams\/its\/blog\/2017\/09\/web-analytics-how-we-collect-data\/","title":{"rendered":"Web analytics: how we collect data"},"content":{"rendered":"\n
\"\"<\/figure><\/div>\n\n\n\n

In this post on web analytics, we will share our experience and explore various aspects of how we in the Web Development team address client-side data collection and analysis. We will also provide a set of recommendations that help tackle common issues and present some use-cases that highlight specific implementation challenges and successes.<\/p>\n\n\n\n\n\n\n\n

Understanding the context<\/h3>\n\n\n\n

“Often numbers don’t speak as loudly as they should because you are missing one simple ingredient: context.”<\/em><\/p>

Avinash Kaushik, author of Web Analytics 2.0<\/em><\/cite><\/p><\/blockquote>\n\n\n\n

Analytics data do not answer questions per se, but are relevant when they stand at the intersection of business objectives, strategic goals and contextual metrics and KPIs <\/a>. Actually, they become <\/em>relevant when they map to outcomes in that they fit into the organisation’s overall objectives and with specific goals in mind, accompanied with accurate measurements, performance targets and aimed at serving identifiable user segments.<\/p>\n\n\n\n

A common problem with data collection is posing the right question about data analysis and interpretation: what does it mean for our organisation to have one or two million hits per month? Is that too many? Too few? Just enough? Also, when do collected data start to provide most of their value? In the future, today or in the past? And who benefits from all of that of data when they are not translated into meaningful sources of knowledge? More importantly, the same numbers might signify different things to different people. Our focus is to put numbers in the context of the problem we are trying to solve, or the outcome we want to generate for the users of our services.<\/p>\n\n\n\n

How EMBL-EBI collects data<\/h3>\n\n\n\n

As a general rule, data collection at the EMBL-EBI is performed in two ways:<\/p>\n\n\n\n

1. Server-side log analysis<\/h4>\n\n\n\n
\"\"<\/figure><\/div>\n\n\n\n

Server-side logs are managed by the EMBL-EBI Web Production team<\/a>. Their main focus regarding data collection is around baseline statistics across all of EMBL-EBI’s online presence. This exercise of data gathering adds to a collection of other indicators of general use and impact. These are used for various executive, administrative and grant supporting purposes. The web log reports produced at EMBL-EBI are based on the logs generated by the web load-balancers. These record the traffic going to the EMBL-EBI hosted domains and associated IP addresses.<\/p>\n\n\n\n

The solutions adopted are the result of architectural decisions, based around the following criteria:<\/p>\n\n\n\n