Undercover UX Research

When it comes to research, budgets often fall shorter than aspirations. But real data can still be used to guide the design process, even when traditional methods are off the table.

Picture the scene. Several project stakeholders are drawn into a heated debate over a homepage design. Everyone has their own opinions and preferences about what is important – and yet nobody can agree.

One of the team members points out that we aren’t our users. “Hey, why don’t we invest a few day’s effort into some user research? That way, we’ll learn what people actually want!” they suggest.

The project lead rejects the idea. “That would be nice,” they say, “but we just don’t have the budget for any research. Let’s just rely on best practice”.

Many of you will find this scenario eerily familiar. The need for organisations to stay competitive and release products quickly can force teams to compress budgets. Much to the chagrin of designers, user research is often the first activity to be cut.

To assume makes an ass out of u and me

The problem with cutting research is that it can result in usability disasters. With no data to guide decision making, we fall back on our assumptions. And assumption is the enemy of good design. To assume is to make decisions based on personal preferences, instead of our target audience’s.

It’s times like these we need to go undercover. To practice faster and less formal research alternatives. In this post, we’ll discuss some low-cost activities that don’t need to be budgeted or signed-off. 

Let’s go back to that initial scenario. The issue was that we couldn’t answer a specific question : “What should we put on our home page?”. A question like this is the perfect frame for a research topic. After all, a home page should be reflective of high-priority user needs. It should highlight the tasks for which a person is most likely to visit. If we can answer this question with real data it will help put a stop to unproductive debates.

A fully scoped and budgeted research phase is certainly nice, but it isn’t the only way.

1. Love your analytics data

Though often seen as a tool for marketers, Google Analytics is one of the most valuable user research tools at your disposal. It provides a huge array of information about site visitors and their behaviours. Best of all, it costs nothing and is immediately accessible. 

The data can be used to answer questions like:

  • What are our users’ interests?
  • Which content & features are used most often?
  • How do users move through the site?
  • Do behaviours vary between devices, locations or demographics?

Although it is possible to go deep with analytics, it can also present some amazing insights at just a quick glance. There’s no need to get sign-off or budget approval for that – it’s ready and waiting to be analysed. In our experience many clients already have Google Analytics set up, even if they have never looked at it. 

Other free analytics tools such as page load speed calculators and accessibility checkers are also incredibly handy. They help us understand the existing site performance, and set goals for improving these factors in our redesign work.

Google Analytics Screenshot
Google Analytics can be a treasure trove of immediate user insights.

2. Turn up the heat

Whilst Google Analytics is great for determining the ‘what?’, it is less effective at telling us ‘why?’. Heat mapping services such as HotJar and ClickTale track actions like mouse-movement, clicking and scrolling at the page level. These tools are a fast and affordable way to get behavioural insight, and are a natural next-stop for conducting some undercover research.

We can use these tools to answer questions like:

  • Are people noticing the important information on a page?
  • What are they clicking on?
  • Do people scroll on long pages?

These tools are awesome because they replicate the effects of eye tracking, but with a much lower barrier to entry (think £50 instead of £10,000). Stakeholders respond well to these heat maps, as they show behaviours in such a visual way. They effectively answer the eternal question “did the user look at what we wanted them to see?”.

Screenshot of heatmaps
Heat mapping tools like HotJar can recreate the effect of eye tracking, at a fraction of the cost.

3. Quick and dirty user testing

Analytics tools are great for gathering site usage & behavioural data. But what about the more emotive side of user research? What if we need feedback on a concept – such as an early design, or a prototype? 

Sometimes we need more human feedback on our designs, not just stats. To find out how our products make people feel, we turn to user testing. Essentially, user testing involves observing a person actually using our product to complete a few tasks, and giving feedback as they go.

There’s a misconception that user testing always needs to be a formal process, run in a lab-like environment under strictly moderated conditions. That works when we have the budget, but when research time is squeezed that kind of approach isn’t feasible. Luckily, the most valuable insights can often be gathered from very casual testing – and can even be done remotely.

Just having a few acquaintances or colleagues play with your website or prototype for 15 minutes can validate the design, and determine whether things are working as they should. You don’t need to make a meal of it. Just provide them some background and a basic scenario, encouraging them to ‘think aloud’ with their feedback.

User Testing
User Testing doesn't need to be a complicated operation. Keeping it casual also keeps things affordable.

There are many free screen-sharing and recording tools available. Almost every decent video chat application has a screen-sharing feature. Macs also come bundled with Quicktime for screen recording as standard. Chances are, your machine already has everything it needs to run and record a session. It has never been easier to share a website link or a prototype, and observe someone playing around on it. 

This sort of testing can be moderated or unmoderated. Of course it’s always preferable to run the show yourself – so that you can ask follow-up questions and make sure the tester ‘thinks aloud’. That said, if you are time strapped just emailing the scenario and a link for someone to look at in their own time will still be really helpful.

Tools such as can make this process a little easier by bundling everything into one link you can distribute to volunteers. It comes at a small cost, but can be helpful if you’re dealing with a less tech-savvy audience who might struggle to record their own screens.

4. ‘Quickfire’ online testing tools

Sometimes we just need feedback on something really specific, really quickly. That might involve getting fast thoughts on a mock-up, or checking to see if a piece of wording is clear to users.

For these kinds of research scenarios, there are a few online services dedicated towards short and targeted testing. Currently our preferred service is UsabilityHub. They give you a whole bunch of test types – including the ‘five second test’. It’s a technique that simulates the first few moments of a person visiting a site, following up with specific questions about what stood out. This is great for testing landing pages, to see if the messaging resonated with people. They also offer the ‘preference test’, where participants are simply asked which version of a design they prefer and why.

Because of the shorter and more focused nature of the tests, it’s possible to get qualitative feedback really quickly, without needing to schedule or analyse fully recorded user testing sessions. These types of tools are great for validating smaller details of the design, such as :

  • Do visitors like the brand look-and-feel?
  • Are landing pages catching the user’s attention?
  • Do people understand the value proposition?

It has never been easier to share a website link or a prototype, and observe someone playing around on it.

Preference Test screenshot
Tools like the preference test can be used to get fast feedback on something really specific.

Remember, all research is good research

There’s no single ‘right way’ to approach research… but any is better than none. Sometimes we just need quick answers, and that can mean going undercover. A fully scoped and budgeted research phase is certainly nice, but it isn’t the only way. 

If budgets don’t permit the ideal scenario, we adapt our approach. In the immortal words of Jakob Nielsen: “Leaving the user out is not an option.”

Share this post

Related Posts