Quantifying “Pogo-Sticking” with Sequence Segments

Man on Pogo Stick“Pogo-stick” behaviour, so named by Jared Spool of User Interface Engineering (UIE) over 15 years ago [1], refers to website visitors jumping around and retracing their steps on a website.  [ You can hear Jared discuss this in an excerpt of his UIE 17 conference keynote. ]

This is a behaviour you might observe in a usability study, or suspect is occurring due to customer satisfaction/feedback surveys.

One of the difficulties of usability study findings is that the data tends to be qualitative.

With competing priorities a given, is there a way to quantify how big the problem is?  Indeed there is, using a Google Analytics Sequence Segment.

When Google rolled out the new Unified Segments in July 2013, sequence segments were added, which allows you to segment your data to find all visits that include a specific path. Therefore, you can use sequence segments to quantify pogo-sticking behaviour.

And even better, after implementing a design change, use the sequence segment to measure the decline in pogo-sticking behaviour, quantifying improvement by measuring actual change in behaviour.  How cool is that?

Here’s how you can create a sequence segment.

Create a Sequence Segment to Measure Home Page Pogo-Sticking

Let’s say you suspect that the copy within specific widget on the home page is misleading.  You’ve taken a look at the Navigation Summary report for the destination page (let’s call it /section-page.aspx), and it does appear that there might be some back button behaviour happening, but you’re not sure.  You would like to quantify this behaviour…

Pogo Stick Behaviour

In Google Analytics, click to open the Segments builder and click on “Create a New Segment”.  Next, click on “Sequences” as shown in the image below, and click on “Add Step” to add the steps that represent the pogo-stick sequence in the image above.

Google Analytics Sequence SegmentBecause we’re interested in clicks that follow in exactly this sequence, we’ve clicked on the “is immediately followed by…” option:

Google Analytics Sequence Segment

Measure Pogo-Sticking Followed by Visit Exit

If you are concerned that the problem might be severe enough to cause the visitor to leave,    you can use the dimension “exit page” for “step 3”.

Google Analytics Sequence SegmentEvaluate Highest One-Step Pogo-Sticking Behaviour from the Home Page

If you want to find out which clicks from the home page result in a shallow dip down and then an exit, use a wild card regular expression for Step 2.  Use the dimension “Landing Page” for step 1.

Google Analytics Sequence Segment

Not limited to just evaluating pogo-sticking, use sequence segments to quantify any non-linear path behaviour through your website.  Measure behaviour before and after website changes, to quantify the benefit of design changes.

====

[1] “Pogo-Sticking: Downs and Ups”, Eye for Design, User Interface Engineering (UIE) newsletter, July/August 1997,

6 thoughts on “Quantifying “Pogo-Sticking” with Sequence Segments

  1. Excellent post. I was checking continuously this blog
    and I am impressed! Extremely useful information specifically the last part 🙂 I care for such info much.

    I was seeking this particular info for a very long
    time. Thank you and good luck.

  2. I love what you guys are up too. This kind of clever work and exposure!

    Keep up the awesome works guys I’ve you guys to my personal blogroll.

  3. Cool Stuff there is not a lot out there for usage of sequence segments in GA. I tested your segment, do you have more inspiration on what you can do with sequence segments. I feel it is is mighty tool, however most of the things i am testing are not giving out enough data
    Thanks from germany Alexander

    • Hi Alexander,
      Thank you for your feedback.

      We use sequences all the time to answer questions of the format — how many people did A, then B, then C versus A then C then B, and what’s different about them? Different sources of traffic? Different geography? Different time of day?

      A, B and C could be different interactions, or different pages, downloads or videos. And if you set up a “user” instead of “session” sequence segment, the A, B, and C could be in either the same visit or more than one visit.

      You’ve given us a great idea for another blog post!

Leave a Reply

Your email address will not be published. Required fields are marked *