- About Archives
- About SAA
- Careers
- Education
- Publications
- Advocacy
- Membership
Update: The comment period for Version 1 closed on August 22, 2016. Version 2 will be released for comments in January 2017. Archivists and special collections librarians should direct further comments to Task Force co-chairs Amy Schindler amycschindler@gmail.com (SAA) and Christian Dupont christian.dupont@bc.edu (ACRL/RBMS).
Count the total number of Page Views of content published online by the Repository for viewing by Users. Page Views may include, but may not be limited to: the Repository’s website, finding aids, online exhibits, digital collections, or digital objects during the time frame selected by the institution (Daily, Weekly, Monthly, Yearly).
Guidelines for collection:
Page Views of each individual page are added to represent the basic measure.
If possible, exclude “Page Views” by the Repository’s own staff (for example, by filtering out a block of IP addresses assigned to staff workstations).
If possible, exclude reloads of a given page.
Exclude usage statistics derived from social media sites maintained by the Repository; see Social Media Reach.
Applications and Examples:
The application of Page Views would show how strong a Repository’s web presence is. What pages are being viewed most and how does that number trend over time (monthly, yearly)?
Count the number of unique Page Views made by unique Users during a designated time frame selected by the institution (Daily, Weekly, Monthly, Yearly). Count each User once during the selected period regardless of how many Page Views are made during the period and regardless of the visit length.
Guidelines for collection:
If a User visits the same page multiple times during the same session, the view is only counted once.
If possible, exclude the “unique Page Views” of Repository staff.
Applications and Examples:
Track the “unique Page Views” during the defined time period to provide information on the size of your online audience.
Calculate the cumulative time a User spends on an individual page during a time frame selected by the institution (Daily, Weekly, Monthly, Yearly). Session time should only include active viewing by the User. Inactivity of 30 minutes or more by the User should be considered an end to the session and accumulated time should be counted prior to the start of the 30 minutes of inactivity. The total number of sessions conducted by an individual User, the individual session times, and the cumulative time of all sessions should be counted. Record the measure in hours and minutes, hours and fractions of an hour, or minutes.
Guidelines for collection:
A Repository may wish to determine the activity of Users in relationship to the types of materials offered. The time may influence the bandwidth needed to allow Users access to materials to ensure all Users can access materials.
Calculating the session time will allow for detailed reporting on trends over time to evaluate User behavior.
Applications and Examples:
An IP address views an online Exhibition and stays active on the page for an hour and then is inactive for 15 minutes, but becomes active again for another hour. The total session time for that IP address would be 2 hours.
Session times during times of high activity could affect all Users who need access to materials.
Determine and count what links the User is clicking to arrive at the various webpages related to the Repository and its online content. This includes direct traffic (clicks from bookmarks or visitors who know the URL), Web searches, links from external sites navigating a User to the Repository, and internal links transferring Users from one part of an organization’s or Repository’s site to another.
Guidelines for collection:
Calculate the traffic source through tally or percentage of Users on the site arriving through that source.
Percentage would be calculated by dividing the number of Users from a traffic source divided by the total number of Users tallied during the corresponding time period.
Application and examples:
Incoming traffic sources could point to successful outreach initiatives using external sites such as adding a link from a Wikipedia entry to related digitized material or a finding aid.
Traffic source could be used as a situational measure. For example it could be used following an online exhibit to determine how Users were finding their way to the content.
Count the number of times a digital object was downloaded during a selected time period (daily/weekly/monthly/yearly). This would include, for example, if a digital image was downloaded multiple times by the same User during the selected timeframe.
Guidelines for collection:
Tally the number of times a digital object was downloaded.
Application and examples:
Counting downloads offers another measurement of usage of Collection Units.
By counting the number of times a digital object is downloaded, the Repository may be better able to select materials of interest to its Users by providing a data point in the prioritization of materials to make available online.
Create categories of digital object types and count the total times each object type is downloaded by all Users during a 24 hour cycle starting and ending at 12 midnight.
Guidelines for collection:
Categories may include, but are not limited to textual, image, moving image, and audio files.
Application and examples:
Counting items downloaded will allow the Repository to better select materials of interest to its Users and provide a method to prioritize and allocate resources to make these resources available online.
If a User downloads a set of pages during one 24 hours cycle (as defined by the Repository); if the same User downloads pages in the next cycle then they would counted again. A 24 hour cycle could be from closing time day to closing time the next day; opening time to opening time the next day; 12 midnight to 12 midnight.
Count the total number of users who follow or friend any of the Repository’s social media accounts.
Guidelines for collection:
Count the number of interactions with each social media resource. Report the type of interaction counted.
Repositories may wish to keep the counts for each resource separate in order to facilitate longitudinal comparisons and evaluations of the distinct reach and impact of each resource.
Interactions types may include number of followers, likes, comments, shares, retweets, etc.
Application and examples:
A Repository that maintains a Facebook page and two or more Twitter accounts may wish to tally separately the number of Facebook likes while tallying and combining the total number of Twitter followers across all Twitter accounts.
A Repository that maintains a Twitter account may wish to count the number of tweets that it publishes, as well as retweets of its tweets by others in addition to the number of followers it has.
Total Page Views per day/week/month/year
Graphing the total number of Page Views per day over a given period of time can reveal usage patterns. For instance, at academic institutions, total daily Page Views might increase towards the end of the semester, when research papers are due. For all institutions, total daily Page Views may increase for specific projects, during specific business cycles, or at times of the year.
Comparing the total number of Page Views per day/week/month/year for multiple years in succession can reveal fluctuations in usage levels and trends.
Percentage of single page visits over some period of time (bounce rate)
Comparing the total number pages to single page visits may indicate the types of information being reviewed or may indicate accidental visits to the site.
Percentage of return visits over some period of time (loyalty)
Reviewing the number of return visit may indicate the interest in materials on the site and determine whether there may be potential demand to add additional related materials to the site.
Average session time per day/week/month/year
Determining session time may reveal the value of the content on the site.
Determining average session time will indicate if there is enough bandwidth for Users to access the materials.
Generally anything over 60 seconds is seen as commitment although Eric Peterson (2004) has argued that visitors who spend less than 90 seconds are uninterested. (Commitment)
Total session time per day/week/month/year
Collecting the total session time for a specific time period will indicate the value of the content and may indicate if additional or related materials should be added to the site.
Determining total session time for a specific time period will indicate if there is enough bandwidth for Users to access the materials. This maybe useful during high activity times such as term end, project end, end of year activities.
Number of unique Users per day/week/month/year
Each unique individual that visits your site within a measured period. A visitor who returns to your site within the measured period is only counted once.
Percent of new User sessions per day/week/month/year
The percentage of new visitors to view your website content within a measured period.
Items downloaded per day/week/month/year
Calculating how many digital objects were downloaded over a measured period. A Repository may also wish to distinguish the file types of downloaded material.
Average items downloaded per User
After calculating the number of unique Users that visited digitized Collection Units and the times a digitized unit was downloaded, determine the average number of downloaded Collection Units per unique User. This metric could show trends in terms of collection popularity and use over days, weeks, months, or years.
Traffic source per day/week/month/year
Track the number of Users being directed to your content from the unique links available to determine any pattern in traffic sources. For instance 40 percent of the Users visiting your latest digitized archival collection may be coming from a link posted on a social media platform. Comparing how traffic is navigating to the content can provide insight for effective future marketing.