Sample Library Custom Reports

Here’s a section from my guide to web analytics that (hopefully) will be finished by my poster session this week. I think custom reports are great for locating some key performance indicators at a glance. Most libraries do not have time to have someone regularly review all of the analytics data, but a day spent creating custom reports can generate more insights than nearly any other effort.

I’ll post the full e-book in a week or so.

That was an artificially brief example with few results (sorry, I’m a small library). However, custom reports have so much potentional, that I created a few generic reports for typical library needs.

Library Outreach

This is report will show how well library outreach efforts are working. In this report, we want to see how long users interact with the site, how much they look at. We want to know how many people we are gaining, where they came from, and what they did.

Metrics: These metrics can be mix and matched. The goal here is to have a variety of metrics about how quickly users began to interact with the page

• Unique Visitors: We will see the raw numbers in the dimensions. However, this will give us a graph. We will want to see a spike of new visitors after our outreach efforts
• Page Views and Pages/Visit: Some how much did first time viewers (or returning viewers) look at. Hopefully this will be high for first time viewers and lower as the number of visits increase.
• Time on Page and Average Time on Page: Another measure of how much users are exploring
• Goals: So what are our outreach goals? View blog? Sign up for program? Request a book? Always have a goal.
• Bounce Rate: Always useful

Dimensions: So we want to segment out users that seem obviously new to the site. Then we want to see where they came from to see which of our outreach efforts are working. For example, if we convinced some different agencies to put a link up to our site from their own, this will tell us which visitors followed that link.

• Visitor Type: This will give us two type: New or Returning. The new visitors will, for the most part, come from the outreach.
• Count of Visits: Drilling down from new visitors, this number will always be 1. However, we can drill down from returning visitors and find the users with low numbers, those will likely be from outreach.
• Source: This will give the domain (such as google.com) and can give an idea of how different users arrived at the site. For example, do users from one particular site come back over and over? That visitors to that site must be interested in what we offer but have not yet been contacted by the libraries. This would be a good place for future outreach.
• Referral Path: This is a little more indepth form of Source. This would be something like localnewspaper.com/community/library This will let us know if they came from a permanent link or from a link from a news story about a library event.

Non-Local Users

For academic and special libraries, the easiest way to separate secondary users (community members, out-of-state users, etc) from the primary user group (students, faculty of the university or department) is to distinguish by geographic region. While this does not cover all users, for example scholars doing field research may be separated geographically, it is a nice, quick way to see how secondary users interact with the website.

Metrics: These metrics can be mix and matched. Here are goal is to see how users intereact with a variety of content. We’ll be drilling down by region and than by landing and exits pages. So for each page we want to see how many users see it, how often its were they land (probably because of search engine results) and how often they leave.

• Bounces: For each page, this will give a quick idea of efficience
• Page Views: How often did users look at a page. This will be an important category as we drill down.
• Entrances/Exits: For each page, how often did users land there and then leave from there
• New Visits/Unique Visits: Do users continue to return to this page? Conversely, does this page continue to attract new users. The numbers can be tricky here. If they are going down, this suggests users like the page and return. If they go up, it may indicate that new users are coming consistently.
• Goals: Always important

Dimensions: First we will sort by geographical area. This will give use a way to segment out our primary users, than we can drill down to see what users like, how they arrive and how they leave.

• City: The top city will almost certainly be the city where the library is located.
• Pages: This will give us a list of content users from each city have viewed. To best understand these results, click on Page Views to sort by the number of times a page has been viewed. Default is alphabetical
• Landing Page: For each page, where did the user start? This is helpful to contrast with page views. Pages with high views are likely what the users wanted, but where did they go first? How long did it take to find it?
• Exit Page: For each landing page, where did the user exit. This will tell us how much more surfing they did from that page or if they, hopefully, found what they wanted and exited.