(This post is part of a series written by the Atlassian technical writers, exploring the latest techniques in technical communication. We’ll write about our projects, experiments and ideas, and we’ll share the techniques we use to give our customers the sweetest documentation in the world.)

As a technical writer, I often wonder what pages people read the most in the documentation I produce and what information they have a hard time discovering. If I can get hold of such information, I can spend more time on the documentation that is of relatively more value to my readers. I set out to get some insight into this by making use of Google Analytics since Google Analytics is enabled for our documentation site.

In this blog post, I’ll share some of my findings from the basic analysis of the Atlassian OnDemand documentation and JIRA 4.4 documentation.

Analyzing the most visited page in Atlassian OnDemand – the Home page

As mentioned in the beginning of the post, I wanted to dig out the pages people read the most. For that purpose, I used Google Analytics to generate a report for the most visited pages in the Atlassian OnDemand documentation of all time since the product launch on October 25, 2011. Once I had this information, I drilled into the visitor behaviour to see what I could find out.

    • I customised the report by showing the most visited pages by returning visitors and new visitors respectively, and there is not a distinct difference in the results of these two groups.
    • Compared with the first round of analysis I did in early November, there have not been any dramatic changes in the top 10 list.
    • The Home page tops the lists this time. The Restricted+Functions+in+Atlassian+OnDemand page is No. 2, although it was the most popular last time.
New visitors' most visited pages
Returning visitors' most visited pages

Where do readers go next from the Home page?

I am also keen to understand readers’ navigation behavior because I can then add or remove content and links accordingly to make information-hunting a easier task for readers.

The following screenshot contains the top 10 links people click after landing on the Atlassian OnDemand home page. Seven of them are links on the Home page itself and the three links marked by red arrows are entries in the left navigation tree.

The most popular 'next pages' from Home

Clearly, the the readers who chose the links in the left navigation tree did not get what they wanted or was relevant to what they wanted on the Home page, and then they went away to look at the navigation tree. If I can discover what they were trying to look for and add it to the Home page, it will hopefully save people a few clicks. With that question in mind, I had a look at the navigation summary for Atlassian+OnDemand+FAQ (marked with the first red arrow in the above screenshot), and the 3 most popular pages my readers went from there are:

The next step for me would be to analyze the navigation summaries for the other two links marked with red arrows.

High-level JIRA statistics

From the following two screenshots of the top ten JIRA 4.4. pages for new visitors and all visitors, we can see that searching is the only user task people read in the documentation besides the introduction pages such as JIRA+101 and installation instructions. And they spend an average of 4 minutes and 11 seconds on this single page, which is a considerable amount of time. As writers, we should add this page to our to-do list and consider investing more time in improving search-related pages as that impacts the most readers.

Most visited JIRA 4.4 pages
New visitors' most visited pages

One other interesting thing I noticed by looking at the above data, is that the average time spent on each documentation page varies a lot from page to page. Generally speaking when someone spends a long time on a page, there are two reasons: 1. The subject is complex; 2. The information is not written well. This led me to wonder whether it would be possible for me to identify the pages that haven’t been written well, by correlating the average time data with other metrics. This will be my next task.

How do you analyze the usage of your documentation?

I’ve only just started playing with Google Analytics, but I’ve already been able to get a rough idea of the pages people find important and gravitate towards. More work needs to be done to find out why, and whether the documentation or the product (or both) needs improving.

Does your team carry out any documentation analysis? Do you use Google Analytics to do it?

We’d love to hear your feedback via the comments section below.

Fresh ideas, announcements, and inspiration for your team, delivered weekly.

Subscribe now