What UX research involves and why it matters for CRO

CRO ( conversion rate optimization) is, in practice, any desired action a user takes on your site. Most often that’s a purchase, but it can just as well be a smaller action that moves the user toward the final decision, like a newsletter sign-up or adding a product to the cart.

UX research (user experience research) is the systematic process of learning about user behavior, needs, motivations and barriers, run for websites, mobile apps and other digital products alike.

Plenty of people associate CRO mainly with quantitative data analysis (GA4 or heatmaps, say). At Digispot we take a broader view, where qualitative research plays an important role too – letting us understand user motivations and barriers.

Later in this article I’ll show which questions quantitative data answers and which qualitative data answers, and why combining them gives the best results. To keep this from getting too theoretical, I’ll draw on specific examples from my own practice.

CRO is concrete, UX research is “fluff”?

Our Digispot clients (mostly e-commerce businesses) usually come to us with a very specific goal – grow conversion. And that’s often the moment CRO gets associated exclusively with numbers, charts and analytics. So when, after an audit, we recommend usability testing, in-depth interviews or other UX research methods, we often see a bit of surprise.

Because for a lot of people, UX research:

• feels “soft”,
• less measurable,
• less directly connected to sales results.

And that’s a serious mistake in thinking.

UX research is a set of methods that let you understand how users interact with a site, and what’s going on in their heads during that process. Through it we learn: what the user understands and what’s unclear to them, where they stop and why, what worries they have before deciding, what frustrates or discourages them.

In short, UX research focuses on improving the user experience, which directly affects conversion.

Turning visitors into buyers happens through a deep understanding of user behavior, intent and needs, and then translating that knowledge into effective solutions.

A CRO audit is excellent at showing what’s happening – it shows the data, identifies where conversion is “leaking”, and helps build hypotheses. But it’s worth remembering that hypotheses, until they’re verified, remain nothing more than assumptions.

Of course there are situations where, based on experience and prior knowledge of users, we can accurately identify potential problems and quickly improve the purchase funnel. Work like that (so-called quick wins) is often relatively easy to ship and can be verified through A/B tests. (A/B tests – how they work and why it’s worth running them)

But from a UX researcher’s perspective, one thing is certain: the real “why” only surfaces through direct contact with users. That matters especially when we don’t have a solid base of qualitative knowledge, like personas or user journeys, that help build sound hypotheses. Every online store has its own characteristics, which is why copying competitors’ solutions or applying generic “best practices” doesn’t always produce the expected results. What works for someone else won’t necessarily work for you (more on this in Adrian’s article: User experience (UX) – how to design for a good user experience).

, say). That’s exactly why qualitative methods – in-depth interviews, contextual inquiry, usability testing – play such an important role. They let you understand the reasons behind the patterns visible in quantitative data (in GA4 or heatmaps, say) and diagnose problems more accurately. They let you genuinely get to know the user: their motivations, barriers and how they make decisions. That lets you build solutions matched to your own business, instead of unthinkingly copying the competition.

Usability testing, for instance, lets you:

• identify friction points on the purchase path that don’t show up in analytics (because the right events aren’t being tracked, say),
• understand user behavior in context,
• verify whether the changes you shipped are actually working.

They’re especially useful where A/B testing isn’t possible (with low traffic, say) or where the changes are complex, as with a full site redesign.

UX research is part of the foundation of effective CRO, because it turns hypotheses into understanding, supplies the “why” behind the data, and lets you design sound UX solutions and better experiments.

Examples of finding the “why”, and a situation where qualitative research uncovered a problem later confirmed by the data.

On one project for our client MyBed we ran extensive quantitative research, including a full audit and in-depth data analysis. On that basis we began optimizing conversion through specific changes to the store’s design and improvements to the user experience.

But it was only the qualitative research that let us see the fuller picture. During testing I noticed user behaviors that either:

• explained the “why” behind what we’d already seen in the quantitative data,
• or completely undermined our hypotheses and supplied new context,
• or revealed entirely new problems, which we then confirmed in the data.

Example 1: “Working” tabs that were actually misleading

Our analyst ran an analysis of the tabs on the product listing. We checked whether users were using them and whether they were doing their job.

The result (in short): yes, the tabs get used, so you could conclude they work.

But usability testing showed something completely different:

• users were clicking the tabs, but mainly becausethey didn’t understand their labelsand were trying to use them to “figure out” the differences between products,
• on top of that, they expected the tabs to work like filters, not the way they’d actually been designed.

Conclusion: the quantitative data wasn’t wrong, but interpreting it that way could lead to the wrong conclusion. The mere fact of clicks doesn’t mean a solution is working as the user intends.

Example 2: A CTA almost nobody used

During the same study I noticed a certain pattern in user behavior. But I couldn’t draw a conclusion based on the behavior of 10 users. I asked our analyst to check it against the data. The data showed that:

• onlyone user in five (18%)started using the main bed-configuration feature by clicking the primary “Configure and buy” CTA,
• while as many as82% of users discovered the feature only while scrolling the page.

That meant:

• the main CTA was practicallyfailing at its job,
• users were landing on the feature “by accident”, not through the path it was designed for.

That insight turned out to be critical for the planned changes. The new, improved bed-configuration feature was meant to be reachable only by clicking the CTA, unlike the previous version, which could also be discovered by scrolling.

Thanks to that discovery, we could reactbefore the change reached production,avoid the risk of misjudging the new feature’s effectiveness, and design a better entry point to an important part of the experience.

These two examples show clearly that:

• quantitative data tells uswhat,
• is happening,but only qualitative research shows.

why

And it’s only combining these two perspectives that lets you make sound design and business decisions.

Summary – quantitative data tells you “where and what to fix”, UX research shows you “why and how”.If real results matter to you – better design, higher user satisfaction, greater retention and, above all, growing conversion – the answer is simple: UX research is a continuous process that shouldn’t end. The benefits of UX research have been documented and confirmed in practice many times over (in research by the Baymard Institute, among others). But it’s only in real projects that you see how much value it brings to CRO work.

Qualitative research delivers something quantitative data alone will never give you: context. Through it:

• you stop guessing and start understanding user behavior,
• you discover the “why” behind the data,
• you build better hypotheses for testing,
• you cut down on random, costly experiments.

Relying on quantitative data alone can lead to guesswork, and UX research on its own, without support from numerical data, won’t show you which problems have the biggest effect on business results. That’s why it’s important to combine both approaches.

At Digispot we take a broader view of CRO – not just quantitative data, but a process built on two key workstreams: research (covering both UX research and analytics, qualitative and quantitative) and experimentation. Put differently, CRO includes both UXR and analytics, which together drive effective tests and real results.

It’s exactly this crossing of quantitative and qualitative research that produces changes that aren’t random but effective, because they’re grounded in real insights.