We’ve joined the Carnegie team! Find out more.
Alert Close close
Intelligence
Four Steps to Better Web Analytics and Measurement

Intelligence

Four Steps to Better Web Analytics and Measurement

Aug 20, 2015By Greg Zguta

Looking to step up your website analytics game? Let’s talk about four things you can do immediately:

1. Plan an Account Structure to Track Your Properties.

Institutions of all sizes benefit from planning the websites they need to track and the Google Analytics accounts to be used. It is often not enough to put Google Analytics code on all pages of the site. It’s a challenge to account for multiple subdomains, school and college sites, and web managers across a campus with separate reporting needs, especially for larger institutions. If you already have multiple Google Analytics accounts in use throughout your institution, you have an opportunity to help web managers focus on the data for their area of the site.

At Fordham University, for example, web managers are interested in traffic to Fordham sites as a whole, but administrators of the university’s newsroom want to focus on goals for that property separately.

The two sites were set up with separate subdomains and use different CMS platforms. Visitors move between the sites seamlessly, because content is syndicated from one to the other, and the newsroom content is tracked in both accounts.

Planning the architecture for the accounts, properties, and views within Google Analytics, as well as how the tracking code will be managed, leads to more successful implementations.

2. Track All Your Data.

Implementing Universal Analytics tracking code across the website is fundamental to healthy analytics. Universal Analytics is the latest version of the Google Analytics tracking code, but many institutions still need to upgrade from the older javascript versions (urchin.js and ga.js, for instance) versions of the code to this latest version.

As always, you want tracking code on every page of your site. Use your content management system (CMS) to help manage this code by placing it within your page templates and make sure it works. A great tool to help spot check pages of your site is the Tag Assistant extension for Chrome. Avoiding gaps and missing data preserves your ability to analyze in the future. Common tasks such as comparing similar time periods are important for higher ed because of the cycles of web traffic around academic calendars and admission seasons.

3. Use Google Tag Manager — It’s the Future.

Fordham University and other successful institutions are using Google Tag Manager to make it easier to deploy tracking code across the site. Instead of putting Google Analytics code on all pages of the site, you place Tag Manager code on the pages instead. Then you use the web-based Tag Manager interface to publish tracking code to your site.  Tag Manager handles tracking code beyond Google Analytics, including AdRoll, DoubleClick, and CrazyEgg, for example, and it can enable more advanced Google Analytics features such as event tracking without the need to put javascript code into web pages or CMS. Web managers can change the tracking code based on changing marketing campaign priorities and enable more sophisticated tracking of multiple sites across multiple systems.

Implementing Tag Manager requires a little coordination and planning, but it is the future of how analytics code will be managed.

The steps:

  1. Establish an account.
  2. Add the Tag Manager container HTML snippet as you would Google Analytics code.
  3. Configure Tag Manager with tracking tags, which are then published to your site. You also can configure Google Analytics event tracking.
  4. Remove the “old” Google Analytics tracking code once the Tag Manager tracking is published.

4. Set Goals and Measure.

Take the time to set goals and decide which metrics are important to your institution. Then consider what you can do to improve these metrics, and be sure you have a way to measure and analyze the results.

Let’s use Palo Alto University (PAU) as an example. The strategy team identified these goals for PAU’s website redesign project:

  • Increase engagement (pages per session and visit duration) for external visitors.
  • Increase mobile and tablet traffic.
  • Evaluate the effectiveness of a new menu of audience gateway page links.
  • Increase the number of unique visitors.

Changes in information architecture, content development, and design were implemented, all focused on meeting these goals. Analyzing analytics for comparable periods before and six months after site launch, PAU found the following:

  • Engagement increased more than 50 percent.
  • Mobile and tablet traffic increased substantially, primarily in mobile.
  • New menus performed well at directing audiences, particularly internal, to resources.
  • Unique visitors to the site increased, but not as much as was desired.

Setting specific goals and analyzing the right data enabled PAU to see what was successful. The institution is now positioned to continue successful practices, develop new tactics to help meet original goals, and set new goals for the website.

Carry it Forward.

Now’s the perfect time to revisit your analytics implementation.Think about your analytics architecture and how you can prepare for healthy data collection. Consider goals tied to institutional objectives that you can set to measure the success of your website. Make sure you are tracking the things that matter, using the latest tools available. And finally, plan to revisit your metrics later this year to evaluate results, develop new tactics to continue improvement, and measure your success.

Need help? Have a success or tactic to share? I’d always love to hear about it via comments, email, or social media.


  • Greg Zguta Director of Web Strategy I've been working on education web projects since the late 90's and enjoy visiting campuses and watching how technology has transformed higher education since I got my first email account at Oberlin College in 1992. Back then, I mostly used the web to check weather radar and sports scores . . . I suppose technology hasn't transformed everything yet.