Interpreting and Using Analytics Data

Web analytics is about more than just collecting data. Learning how to track information about how users interact with your website, and how to interpret this data, can help you make informed decisions about your content. Using this data effectively will allow you to make content and structural refinements, present better reports to your leadership, and keep your digital platforms aligned.

Interpreting and Using Analytics Data

Site analytics and Siteimprove 

Yale University offers website owners access to Siteimprove, a powerful and user-friendly analytics and optimization platform. Some common metrics include: 

  • Pageviews: Numbers of times a page was visited. 
  • Visitors: User demographics, such as country, device, browser and operating system. 
  • Bounce rate: Percentage of visitors to a particular website who navigate away from the site after viewing only one page. 
  • Exits and entries: Entry rate refers to the percentage of visitors who had this page as an entry page. Exit rate is the percentage of visitors who clicked a link to another website, either internal or external.  
  • Traffic sources: Shows you how your users get to your site. The four main sources are: search, direct, external, or social. 

To get started with Siteimprove, send an email to yalesites@yale.edu with the site you would like access to by providing the name and URL

Siteimprove Training Resources 

Siteimprove Help Center Support & FAQs 

Defining your goals  

With so much information at your fingertips, it’s important to establish what you want to achieve with your data—and how you’re going to do it. Our Interpreting Data and Analytics webinar will help you: 

  • Establish a fundamental understanding of common web metrics 
  • Define goals and translate those into a meaningful measurement plan 
  • Use your data to revise and optimize your site content 

Adding context to data 

What do you do when your leadership asks for numbers, or wants to see how something is performing?  

Numbers without context miss a major opportunity: to tell a story that helps the viewer of the data understand why they’re relevant. Whether the data look good or look alarming, think about the additional metrics or context you might want to cross-check.  

Example 1

You see a large spike in traffic in late May every year, but this year, it’s missing. When you look at your monthly stats, May looks similar to April but very different from May last year. Viewing monthly stats alone might not tell the story. Nor would year-to-year. But if you know that streaming commencement went from being embedded on your site to being hosted on YouTube and linked directly via email, this drop in traffic is easily explained.  

As you prepare reports, ask yourself: 

  • What is the story you want to tell? 
  • What context will be helpful? 
  • What sort of learning is your leader(ship team)? 

Also, beware of vanity metrics—data that appear to be meaningful or impressive, but don’t give insight into the true performance of your digital property.  

Example 2

A page on your site received a 100% increase in traffic (Pageviews) in April compared to March. At face value, this increase in traffic is impressive. Your content is getting a lot of views! However, it may become less noteworthy with additional context: 

  • You become aware of a widely distributed email that erroneously provided a direct link to the page. 
  • Most traffic comes from an international location that is not a known primary source of traffic. This could indicate that your page is receiving bot traffic.    

To avoid vanity metrics, ensure you’re selecting data that: 

  • Contain the context needed to make a noteworthy comparison, and/or; 
  • Measure an aspect of the page that corresponds to an organizational goal. 

Improving content based on data 

Once you’ve gathered and analyzed data, what do you do to improve your content? You might:

  • Examine search queries (an indication of what people are searching for and not finding on your site) and revise your headers and subheadings to add relevant keywords. 
  • Look at low-traffic pages and consider what you might want to change about the content or placement in the site’s architecture, or whether the pages should exist at all.  
  • If behavioral maps show that visitors aren’t scrolling deep enough to find your calls to action, consider moving them higher on the page.

Join us at our next live training

Attend an upcoming training on interpreting data and analytics.

Upcoming Trainings