One of our initial activities at the beginning of the discovery phase of our website redesign was to analyze the numbers behind our visitors and observe their activity on the website. We reviewed Google Analytics for Fiscal Year 2016 (June 1, 2015 to May 31, 2016) to get a recent snapshot, as well as Google Search Console[1] which allowed us to see top search terms.

Not only did we look at typical metrics such as number of visitors and sessions[2] but we closely examined all available data included in these areas:

  • Audience: 95% of sessions came from the United States with most traffic (65%) originating locally from the domain. This means most traffic is coming from on campus. Our users are primarily English-speaking. Only 1% of users have their computers/devices set to Spanish.
  • Access: Most traffic to the site was direct (63%), meaning our users were directly typing our URL in the browser. Our second source of traffic was through organic searches[3] in Google, Bing and Yahoo using a combination of these words: um, library, miami, university of miami, and richter. Referrals[8] accounted for 16% of traffic, mainly coming from the library website itself, the campus website, the old catalog, Summon and SubjectsPlus.
  • Top Content: In addition to main home page, our top pages included home pages for other branch libraries/collections, hours, books, articles and the databases page. Our top 2 databases were PsycINFO and JSTOR.
  • Engagement: More than a third of visits to our website were to a single page. This could mean either users got what they wanted right away, or they had the library home page set as the default on their browsers (as in the commons area or other public machines). 75% of sessions lasted between 0 and 10 seconds. Users who visited at least 2 pages per session spent an average of 3 min 55 sec on these pages.
  • Technology: Desktops still dominated all devices used in sessions; however, there was a 15% increase in mobile sessions compared to Fiscal Year 2015. There were approximately 4K new mobile users while the use of tablets went down by 27%. The majority of sessions came from Chrome browsers on Windows machines. The use of Firefox dropped by 6% while Safari use was consistent.
  • Page Speed: Our top pages had an average page load time of 3.96 sec. Using additional tools ( and Google’s PageSpeed Insight) we were able to compare our performance against those of our 10 peer institutions. We ranked 4th for mobile speed, 2nd for mobile user experience and 8th for desktop speed.
  • Custom Events: Custom event monitoring helped us identify the most accessed areas in the header (links back to the homepage and hours). The most accessed navigational items were books, patron info, research guides and booking study rooms. The top “Quick Links” clicks were for WorldCat, booking study rooms, and the PsycINFO database.
  • Social Referrals: Top social referrals came from Facebook (61% of sessions) and Twitter (12% of sessions). Pages with most social traffic from Facebook were the main website, Cuban Heritage Collection (CHC) and Special Collections. Pages with most social traffic from Twitter were the music website and CHC.

Glossary Terms

  1. Google Search Console: (formerly known as Webmaster Tools) a comprehensive set of tools for in-depth monitoring of various aspects of website’s appearance and performance
  2. Sessions: group of interactions that take place on your website within a given time frame
  3. Organic search: a search originating in a search engine like Google, Bing, etc.
  4. Responsive web design: approach that suggests that design and development of a website should respond to the user’s behavior and environment based on screen size, platform and orientation
  5. Viewport: the visible area displayed on a screen or mobile device
  6. Cached: temporary storage of HTML pages and images to reduce bandwidth use
  7. Bounce rate: the percentage of visitors to your website who navigate away after viewing only one page. The lower the bounce rate the better.
  8. Referral: Traffic to your site that come from direct links on other websites rather than directly or from search engines

The insights we obtained from our Google Analytics data helped us draft technical recommendations and suggestions for the interface design as it pertains to our content.


  • Deliver a responsive web[4] experience that meets our increasing number of mobile users
  • Have 2 main viewports[5]: 768px and under for mobile, 1024px and up for desktop
  • Support browsers at least 2-years old at the time of website redesign
  • Optimize website speed and performance by minimizing file sizes
  • Serve a cached[6] version of the website when possible to allow for speedy loading times


  • Include easy access to key pages from the home page and main navigation menu
  • Include easy access to top databases from the home page or databases page
  • Offer more engaging interactions with content so there is more activity in a page – to lower bounce rates[7] and increase time spent on a page
  • If we want to promote our social channels perhaps add visibility to these elements; the same applies to our news

In our ongoing efforts to maintain best practices we will continue to monitor our referrals and organic visits, as well as continue to run page speed testing tools to monitor and improve performance. We will share key metrics with departments across the library by setting up Google Analytics dashboards/custom reports for each branch/collection website, and weekly/monthly auto sending emails/reports to appropriate people in branches/collections/departments.

Curious about our numbers? See the complete Google Analytics Discovery Report.