UIEtips: Activity Analytics for Web Apps

Jared Spool

November 15th, 2012

In today’s UIEtips, I talk about analytic tools and the insights of what they can capture.

Here’s an excerpt from the article

Tools like Google Analytics or Webtrends are very powerful for content sites, where the only things a user does is visit a page, click on another page, or leave the site. These tools can deliver insights on which pages people visited and in what order.

However, web-based applications are more complex. When implemented with something like a model/view/controller approach, it’s not uncommon for one “page” to serve many purposes. A single page can change dynamically depending on the view and model involved, even though its the same page. In a traditional analytics tool, that one page will show up inaccurately, without giving any insight on about what actually is going on.

Similarly, these tools can’t detect non-page interactions, such as lightboxes or pop-up dialogs. Sophisticated user interactions go unreported.

The result is the team can’t trust the reports from the tool. Once the team loses confidence in their reporting data, its value is greatly diminished. A different approach is necessary.

Read the article: Activity Analytics for Web Apps.

Find out more about the retailer that increased their revenues by $300 million.

Do you use Page-based Analytics? Or have you implemented your own Analytical system? What insights have you gained?

7 Responses to “UIEtips: Activity Analytics for Web Apps”

  1. Michael Whitaker Says:

    Hi there,

    This statement that a tool like Google Analytics cannot capture non-page interactions is wrong. Google Analytics in particular has a whole tracking methodology called event tracking that is designed for the types of interactions described in the article. We have routinely been tracking error messages in Google Analytics for a long time (eg http://www.michaelwhitaker.com/blog/2009/09/14/surprising-checkout-error-analysis/)


  2. Jan Says:

    This is exactly what we are trying to solve with our start up http://www.uxpro.be It looks at measuring user experience from a task or activity perspective. People do tasks or perform activities and while doing this our system captures what happens: screens displayed, features used, errors made by the user, etc. It gives context to what your users are doing and help you understand why they are doing things. Currently we are in private beta but still are looking for people to help us improve the service. So if you don’t have the datastore maybe we can help.

  3. AJ Says:

    Our IT partners built a phenomenal activity based analytics system. The trouble is building a good mechanism to analyze the data and tell a story. We have the data, and its good, but I am struggling to work with the data at the aggregate. I want to determine things like did a user click on that optional feature and enter data, or just look at it.

  4. Chris Says:

    Similar to Michaels comment, Webtrends tracking can be linked to page level events, even going so far as to have heat maps reporting and real time activity streaming. Additionally, neither GA or Webtrends use server logs for their data.

    Please be careful when you name drop, as your article presents inaccurate information.

  5. Keith Says:

    Tealeaf is really good at providing the inpage insight with forms. Easy to set up events on the fly . Quite pricey but if you have high volumes of traffic using complex forms it will pay back the investment. There may be cheaper tools that will give you 80% of the benefit . Tealeaf not so good on non form pages although heard clicktale may address this issue

  6. Matt de Leon Says:

    Curious if you’ve experimented with analytics services like Mixpanel? It’s incredibly flexible and the daa visualization tools are powerful. I use Mixpanel tp capture everything from errors a user sees to profiling the performance of a piece of code in my app. There are several SaaS tools in the marketplace that solve exactly the problem you’re describing.

  7. Billy Says:

    I’m seriously looking at Klaviyo. Cohort metrics is necessary for subscription-based models.

Add a Comment