Google Analytics Usage Reports for CARLI Digital Collections

Revised: July 5, 2023

In Fall 2015 the Created Content Committee decided to switch methods for collecting Traffic Sources. CARLI will no longer release written reports for Traffic Sources and Keywords, and will instead release the entire dataset in Excel with some initial cleanup and formatting and allow institutions to create reports out of this data that are relevant to them. View Historical Google Analytics Usage Reports for CARLI Digital Collections (FY2011 - FY2015) here.

Due to the nature of our consortial environment and shared CONTENTdm server, we cannot create customized Google Analytics reports for individual collections or institutions.

expand / collapse all

Explanation of the Fields Used in Reports

Source: This field indicates the origin of the referral to the site (i.e., where the visitor came from). Possible values for this field include:

• A domain name ending in .edu, .com, .org, etc. This indicates the web site from which the traffic came. Sometimes this will be your own institution’s web domain, or another web site (e.g., a blog) that linked to your content.
• Google, Yahoo, or another search engine. This means that someone reached the page via a search.
• Facebook, or another social media site. This indicates that visitor came to the page from a link on a social media site, either your own, or another person/institution's.
• Browser Bookmark: This means that the user had previously bookmarked the page in their web browser, and visited the page directly by using that bookmark.
• Tinyurl.com or another URL-shortening service (bit.ly, goo.gl, etc.). These sites shorten long URLs to make them more amenable to Twitter and other sites and services with character or space limitations. When they’re listed as a source, there is no other indication of where this link might have been posted or seen.
• Wikipedia. This indicates that the visitor reached the page through a link on Wikipedia.

Institution Code: This field consists of a three-letter code for each CARLI institution. This should be the same institutional code that is part of your collections’ URLs.

Landing Page: This field identifies the first page that was visited during a particular session. When the visitor clicked on the link, bookmark, or search result indicated in the Source field, this is the page that they came to. This might be browse page for all of your institution’s collections, a URL for a specific collection, or a URL for a single item. A URL may appear multiple times in the statistics if there are multiple sources for that URL.

Sessions: This column tells you how many browsing sessions occurred at a particular landing page, from a particular source. In simpler terms, the Sessions column indicates how many times someone visited a particular page from a particular place. A session might consist of multiple page views over the course of an hour, or a single page view in under 30 seconds. Sessions expire after 30 minutes of inactivity, when a user leaves the site, or at the end of the day.

% New Sessions: This column shows the estimated percentage of sessions came from new users (e.g., users who are visiting for the first time). If the % New Sessions column has a value of 50%, that means that half of the users are returning users who’ve visited the site before, and half are new visitors who’ve never been there before.

Pages/Session: This column shows the average number of pages a user visited a session. A value of 1 in this column means that in the average session, the user left directly after visiting the indicated landing page, and any value higher than 1 means that the user navigated to other pages on the site during the average session. Avg. Session Duration This field shows the average length of a session.

Note: The session-related statistics are for combinations of a particular source and a particular landing page. This means that you can easily compare the behavior of visitors from different sources to the same landing page, for example, but the percentages, especially, are more difficult to accurately compile for a particular page across a number of sources.

If you have any questions, please contact the CARLI Office.