Data Collection – Step 2 of Web Analytics Process

Reading time: 4 minutes

We have seen last week, with our article « Data Measurement – Step 1 of Web Analytics Process » , that it is mandatory to do an initial work of understanding and defining your website objective, in order to set up accurate goals, and build related measurement KPIs. Now that we know which data we need, we can start thinking about Data Collection. This blog post will give you a few insights and advises on how to collect data correctly, following the tools you want to use, and the data you want to collect.

Collecting Basic Data

In 1990, we were doing web analytics by using the log files of our server. This method was good because only you had access to your data on your own server, but the data were way too inaccurate to be really usable. Later, we started doing web analytics by implementing JavaScript scripts from a third party on your web pages, which allowed us to get much more accurate data, but less privacy, since your data was accessed by an external service provider.

Today, we are still using the second method. Actually, this is exactly what Google Analytics (or other similar web analytics providers) is offering to website owners. As one of the best web analytics providers, Google Analytics allow you to collect basic data from your website, such as Bounce Rate, Page Views, Geolocation, Browser Parameters, … by only implementing a Java Script tag given by your web analytics provider on each of your page.

This method remains the one that gives you the most accurate results.

Collecting Specific Data

Web Analytics doesn’t just consist in studying the page views and bounce rate of your landing page. It gets very interesting when you can start studying more specific events, such as a click on a button, the filling of a field, the successful sending of a message on your contact form, etc. The thing is, tools such as Google Analytics do not do this naturally. Indeed, you need to manually implement this events on your website by giving them names, categories, action types etc, and send them to your web analytics providers when you decide to trigger these events

These specific datas are very important to point out micro issues your customer can face, or also to measure the achievement of specific goals.

To conclude, the data collection is pretty easy, since the script based technic to execute the collection is pretty easy to set up. However, there are some data, less generic and more specific to your website or application, that need to be sent from your website, in order to be collected by your web analytics provider. Once you’ve done all of your data collection, you will be able to start analyze them and have a first look at your KPIs.




Stay updated! Join our newsletter for webanalysis tips