Nutrition profile
Extract insights from analytics data and understand the behaviors of website visitors, with help from tools such as Google Analytics and Siteimprove.
Cooking time
Usually under an hour if you're looking to answer a specific question. Otherwise, the time needed depends on the size and complexity of your website data.
If you'd like to export datasets for manual analysis, allow yourself more time.
Perfect for
Learning what users are doing on your website. Analytics will tell you the "what" but not the "why" in user behavior.
Prep work
Know your research questions
Define the questions that you want to answer with the data. For example:
- "What's the most popular page on my website?"
- "How much time do people spend on this page on average?"
- "Which part of this page is the most clicked on?"
Focus your questions on specific parts of the website or certain web pages, and avoid qualitative or evaluative questions like "Do users like my website?" Keep in mind that data are only good at telling you what users do on your website. In order to figure out what they think or why they behave certain ways, you'll likely need to pair analytics data with other research methods, such as usability testing.
Set up an analytics tool
An analytics tool uses trackers (usually segments of JavaScript code) to log user behaviors including their entrance to and exit from your website, the time they spend on each web page, and where they click on a web page. To put trackers on your website, you'll need to set up an analytics tool first.
Google Analytics (free) and Siteimprove (paid) are two popular options on the market. Refer to help articles from Google Analytics or Siteimprove to complete the setup. Reach out to your IT support for help if necessary.
Ingredients
You'll need a web analytics tool, such as Google Analytics (free) or Siteimprove (paid), and have it set up on your website.
The analytics tools should provide you with plenty of insights and visualization with built-in modules. However, in case you'd like to download the raw datasets for manual analysis, you'll also need a tool to log the data and create visualizations. Microsoft Excel usually does the trick.
Directions
We'll use Google Analytics as an example and try to answer some research questions that are often asked. If you're using other tools like Siteimprove, you can find similar features with slightly different naming.
To begin with, don't forget to set a date range for the data to be tracked. In this example, we're tracking the University of Arizona Libraries' web analytics data from December 24, 2019, to December 24, 2020.
Most popular pages: What part of my website do people visit the most?
Go to Behavior → Site Content → All Pages from the sidebar. In this example, the University of Arizona Libraries website's most popular page is its homepage, as with most websites. Other highly-visited pages include those about library accounts, finding articles and books, and visiting library spaces.
Average time on page: How much time do visitors spend on my pages?
There's more to explore under the same view! In a different column "Avg. Time on Page", you'll see the average time people spend on each page of your website. If you're interested in seeing this number for a specific page, put in its URL in the search box near the top of the table.
Remember, this number isn't always the bigger, the better—for example, you shouldn't expect users to spend too much time on a landing page of a program, as you'll often want the page to effectively direct people to more in-depth content.
Bounce rates: Do visitors leave my website after seeing a page, or do they continue to see other pages?
Also under the same view, you'll see a column called "Bounce rate." The term stands for the percentage of your visitors who only visited one page on your website without navigating further to another page. To quickly view data for a specific page, type the page's URL in the search box near the top of the table.
We normally want to see a low bounce rate for web pages because it signals high retention rate for visitors. However, there are exceptions—for example, it's okay for a library hours page showing when the library's open to have a high bounce rate, since it's common for users to navigate to this page, find the information they need, and then leave.
Traffic sources: How do visitors get to my website?
Navigate to Acquisition → All Traffic → Source/Medium, and you'll see a list of sources where your website traffic is from. In our example, 44% of visitors navigated directly to the University of Arizona Libraries' website, while about 39% came from Google and other search engines.
Google Analytics also tracks the amount of traffic your website gets from social media. Under Acquisition in sidebar, navigate to Social → Network Referrals, and you'll see the a list of social networks along with the page views they each contributed.
Plating
Both Google Analytics and Siteimprove provide robust features for generating reports. Aside from taking screenshots from the reports, you can also export them to HTML or PDF files or directly share them with colleagues using a link.
When presenting your findings, you'll often want to include these visualizations in your research report or a slide deck. Consider to also include the following:
- Research questions
- Methods: how you extracted the data. Specify your tools, duration of tracking, pages being tracked, and individual metrics.
- Findings: behavioral patterns revealed by the data. If you're tracking multiple metrics, mention any correlation between them.
- Recommendations: where to go next based on your findings. You may also bring up other research questions that may also be helpful to the topic, and how they can be answered using analytics.
Pro tips
In addition to web analytics, you can also adopt similar methods in tracking data such as online sales or marketing campaigns. For example, Mailchimp (an email marketing service) provides insights of reader participation on each campaign, such as number of clicks on links in an email. Identifying the relationship between these metrics and campaign content can potentially inform improvements to your email strategy.
Some analytics tools such as Siteimprove also have quality assurance features that send you a notification when it detects broken links and misspellings on your website. You can set up accounts for your content managers so that they can promptly fix the problems.