June 15, 2020

UX Analytics: tools, testing and pro tips

By

Theta

“Normal work” has been changing a lot for people working in customer experience! Many UX designers are working from home and might do so for a while. This means you’re probably adjusting the way you collaborate and undertake research - including using Miro boards, real-time ‘co-editing' within your Adobe XD document and Teams/Zoom user interviews.

You’ve possibly started exploring tools such as Google Analytics, Hotjar or Google Tag Manager, which may have even led you to this post. For those who are new to these tools, there’s a lot to learn. To help you get started, I’ll be giving some practical advice on what you should do in terms of setup and analysis, as well as a few pro tips I’ve picked up while working on client projects.

What are your goals?

As a User Experience Designer, you’ll be focussed on improving the experience of a customer or employee on a given app, site, and frequently now, it’s both. You’ll also be looking for ways to validate your input to the business process by demonstrating how your UX/CX work is contributing to your business achieving its goals. Creating goals for these improvements and ways to measure if you’re achieving them is an art form of its own. Google offers a great canvas with its Measurement Plan. See: Plan in Unit 1, Lesson 5 of the Advanced Analytics Course.

In this plan, you’ll find some common metrics to validate that your design changes have helped. Some examples include demonstrating the following:

  • Decreased time to complete a task.
  • Decreased abandonment in a form or process.
  • Decreased interaction with help pages or use of help tutorials.
  • Increased number of downloads or interactions with a white paper.
  • Increased number of newsletters sign up.
  • Increased average session duration or decreased bounce rate.
  • Increased returning user rate or time on site/page.
  • Increased interactions/video plays or completion of contact forms.
  • Increasing scroll depth or content viewing completion.
  • Increasing feedback scores/ratings.

As a designer, you might also have to step outside your analytics platforms to check for wider impacts. These would include a reduction in calls/emails to support teams, feedback forms/NPS systems, interactions with your chatbot or an actual number of sales in a given time period from an external system such a POS. You can bring this data into your core analytics platform for further measurement and analysis.  

Picking the right tools to use

To monitor all of your goals, you need to have the right tools in place. Typically, on an engagement, I start with recommending three tools, Google Analytics (GA), Google Tag Manager (GTM) and Hotjar. There are lots of other tools out there that help with the same thing, but you should always review the policies with your legal team about data, GDPR and privacy before touching any of them. Here are some tips for setting up:

  • Account setup - Make sure you have multiple views set up on your Google Analytics accounts. Once data is modified, you can’t turn back the clock - so keep your raw data in a separate view. An example here would be creating a view that excludes your internal IP address traffic.
  • Values - Extrapolating goal completion or metrics into economic value will help you in your business cases. For example, you might say that a customer completing a feedback form is worth $25 to your business, the equivalent cost to an incentive of a user test or a survey. You can then use these values to create an overall economic assessment of how your design or CX improved the business outcome.
  • Segments - Both Google Analytics and Hotjar allow you to create segments or audiences who meet a certain criterion for analysis. For example, you might only want to review customers from a region or city or those that completed a specific step in a form. These are very powerful and really help to provide value to your customers or colleagues.
  • Planning - Forward planning with custom dimensions and metrics will require some serious thought, and you will likely need to work with an analyst, information architect or data scientist. Crap in, crap out as they say.
  • Test everything - It goes without saying usually, but with analytics, you will want to test everything. A developer can help you set up a Selenium script to automate data entry which will save you time and effort.
  • Sharing the outcome - Once you get your goals set up and your data coming in, one of the most important things you’ll need to do is get the outcome shared with stakeholders and decision-makers. Thankfully, you can set up automated reporting in GA. Or if you’re really fancy, Google Data Studio lets you create a shareable dynamic report styled to your organisation’s branding.

Some pro tips

  • Single Page Apps - Single page applications will need a developer or someone familiar with JavaScript to work on a project with you to get GTM working - your HTML/CSS skills just won’t cut it. Often these systems don’t have an easy way of setting up triggers, and you’ll need to inject code into the DataLayer to get things firing in the way you want.
  • Using the DataLayer - When you get the DataLayer in GTM working how you want, it can provide great context to your analytics reports. It’ll help you understand dynamic behaviour that could otherwise take hours watching Hotjar recordings to make sense of.
  • Cached data - This will ensure your session refreshes, and you see all changes made to the page.
  • Tagged links - Track everything with campaign tags using the Google Campaign URL builder. People are referred from all over the place. The more you embed these links into your work, the happier marketers will be.
  • Site performance - There are limits on how many tags you can fire. The number of scripts you add to your site can also impact performance, so look carefully at this. It can also impact your search rankings which are often critical for businesses. Turn off Hotjar when you don’t need it.

Getting started

If you haven’t used GA or GTM in a while, I would highly recommend reviewing the Analytics Academy free courses and taking the Individual Qualification Exams - all free! But there are countless other free or paid resources out there to help get you started.