Intro to Google Analytics for UX

Bettina Lechner

hello@bettinalechner.com

Omaha Coding Women | 06/22/2015

Web Analytics & UX

  • Traditional research methods: small number of users are observed
    • "What problems do users encounter?"
    • "How do they understand the information presented?"
    • "Can users complete task X quickly?"
  • Web analytics: observe behavior of large numbers of users
    • "Which pages are most/least visited?"
    • "How do mobile users differ from desktop users?"
    • "Is button X more effective in getting users to complete a purchase?"

Google Analytics Basics

setup, terms

Setup

<script>
(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
(i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
})(window,document,'script','//www.google-analytics.com/analytics.js','ga');

ga('create', 'UA-XXXX-Y', 'auto');
ga('send', 'pageview');

</script>

Universal vs Classic

  • Set of improvements to tracking
    • Track users across devices/sessions/etc.
    • Android & iOS
    • Enhanced e-commerce tracking
    • Track links based on ID
  • Upgrade Center

Terms

metrics, dimensions, conversion rate, bounce rate...

Metrics

Numeric measurements of user behavior

  • Avg. time spent on page
  • Pageviews/session
  • Unique visitors
  • % new visits

Dimensions

Categories by which users can be grouped

  • Location
  • Browser
  • New/returning
  • Phone/tablet/desktop
  • Screen size

Visits: instance of a user entering the site, doing something, and leaving

Unique visitors: number of unique visitors within the selected timeframe (based on cookie)

Pageviews: number of times a page is visited (multiple visits in a session are counted multiple times)

Bounce Rate: number of users who entered and left a page without visiting another page

Learning About Users

Visitor Analysis

Visitor Analysis

  • How many use screens smaller than 1280px?
  • How many still use IE < 10?
  • Is it worth it to put a lot of effort into a responsive site?

Useful? Sure – but not that interesting

Some more interesting questions

  • How do users come to the website?
  • What are they really looking for?

Learning How Users Got to the Site

Traffic Analysis

Source & Medium

  • Source: URL or direct
  • Medium: Type of source
    • Organic Search
    • Paid Search
    • Direct
    • Referral
    • Social
    • Email

Accessing Search Keywords

  • How do users find the website?
  • How can we tailor to users' interests?

 

  • Not available within GA anymore (since switch to SSL by default)
  • Accessible via Google Webmaster Tools
  • Only available for the past 90 days

Learning How Content is Used

Content Analysis

Content Analysis

  • High/low pageviews
    • Any pages deep in the hierarchy with high pageviews?
      • Do they have enticing link labels or are they landing pages?
      • Through which medium are users reaching the page?
  • Unique Pageviews much lower than Pageviews?
    • Users may not be finding what they are looking for (returning to the page many times in a sess.)

Content Analysis

  • Low time on page
    • Content not matching expectations
    • Content not interesting
    • Content well organized and questions answered quickly
  • ​High Entrances vs Unique Pageviews
    • Page is often an entry point for users (linked to from another site?)
    • May rethink structure/hierarchy of the page

Content Analysis

  • High Bounce Rate/High % Exit
    • Compare using "Comparison" view
    • Users either find what they're looking for on that one page or don't think your site is relevant
    • May be caused by inappropriate links from other sites
    • What next? Usability test or expert judgment

Content Drilldown

  • Explore pages in hierarchical roll-ups
  • Useful if your site is organized in logical hierarchy
  • View metrics aggregated for each directory

Learning How Users Move Between Pages

Behavior Flow Analysis

Behavior Flow Analysis

  • Complete picture only possible for small number of pages (5-10, maybe)
  • Instead, focus on page-to-page relationships
    • Where did users come from
    • Where did they go
  • Useful in planning usability tests (try to find unexpected pages and then use the test to find out the thought process leading to the behavior)

Behavior Flow Analysis

  • Two different reports for this information:
    • Behavior > Site Content > All Pages > Navigation Summary: Previous and Next Page Path
    • Behavior > Behavior Flow: Flow of users around one specific page (limited usefulness for unguided exploring)

Learning How Groups of Users Differ

User Segmentation

User Segmentation

  • Compare behaviors groups of different users
  • Based on, e.g.
    • internal vs external
    • location/language
    • returning vs first time
    • desktop vs tablet vs phone
    • medium/source

User Segmentation for UX

Did users behave differently if they ...

  • visited page X (e.g. more likely to complete a purchase)
  • entered through page X?
  • entered through a link in an email?
  • entered through a paid ad?
  • visited the careers page?

Advanced Topics

Tracking Custom Actions

ga('send', 'event', 'category', 'action', 'label');
  • Use custom JS to track any user action
  • Useful for tracking outbound destinations and in-page actions
  • Can be linked to conversions/assigned a value
  • Group by category, action, and label
  • More info on developers.google.com

Goals

"Finish line" for a conversion

Conversions

A sequence of actions you want users to take on your website because they provide value to the business

  • Add items to cart > review order > fill in details > submit order
  • Review information > sign up to receive further info

Custom dashboards

  • Metrics by geographic region
  • Compare new designs to old versions
  • Exits and custom events
  • Metrics by directory (e.g. department)

Single-Page Apps

GA    UX

Integrating Google Analytics into the UX portfolio

Personas

(descriptions of archetypical users)

  • Segmenting based on personas
    • May be difficult to find appropriate criteria
    • Be able to answer specific questions by persona, not just overall
  • Building better personas
    • Use analytics data to build more realistic personas based on actions/common paths/characteristics
    • Iterative process of building segments and adjusting them based on results

Usability Testing

(testing an interface on actual users)

  • Test planning
    • Use analytics to identify problematic pages/flows and focus testing on these areas
  • Test analysis
    • Compare usability test results to analytics data (larger user group)
    • Incorporate analytics data into test reports

Usability Inspections

(formal evaluation of interfaces)

  • Test planning
    • Use analytics to identify problematic pages/flows and focus testing on these areas
  • Test analysis
    • Use inspection to determine reason for problems (analytics can only tell you that there is a problem)

Intro to Google Analytics for UX (OCW)

By Bettina Helgenlechner

Intro to Google Analytics for UX (OCW)

Intro to GA and how to incorporate it in your UX portfolio

  • 548