In an era where AI models can write essays, design wireframes, and synthesize market reports in seconds, it’s fair to ask: is user research – especially in loyalty programs – next on the chopping block?
We’ve heard the anxiety. And while the landscape is shifting, our work is far from obsolete. In fact, the rise of AI might just highlight how indispensable human-led UX research really is – especially when it comes to understanding loyalty behaviors that are emotional, contextual, and culturally nuanced.
AI Is Excellent at Answers. UX Research Is About Asking the Right Questions.
Let’s start with a distinction that gets lost in the hype: most AI tools, including large language models (LLMs), are fundamentally retrospective. They draw from existing information, patterns, and language to give you a plausible answer to a query. That’s useful. But it’s not the same as creating new knowledge.
In loyalty program design, many of the core challenges we investigate don’t yet have clear answers. How do first-time users interpret ambiguous point systems? What unspoken mental models are formed after one confusing redemption attempt? Why do certain demographics churn—even when the perceived value is high?
These are not “data problems” waiting to be solved. They are behavioral puzzles rooted in emotion, context, and motivation. You need to get in front of people, understand their lived experience, and decode what isn’t said directly. That’s primary research—and it’s something no LLM can do on its own.
Loyalty Research Lives in the Gaps AI Can’t See
Here at FinUXlab, we’ve audited over 30 loyalty programs in the last year alone—from retail and e-commerce to financial services and mobility. What we consistently find is that the friction points that cause the most damage to loyalty aren’t flagged by dashboards or NPS scores. They live in the gray areas: poorly communicated rules, confusing edge cases, emotional letdowns at the point of reward.
Retail: We conducted an in-depth UX audit of Tesco Clubcard, revealing how inconsistent language between physical receipts and app-based offers was reducing perceived value among older users.
Banking & Finance: Our research with a leading European neobank uncovered that users often abandoned the process of activating cashback rewards due to opaque benefit descriptions hidden behind multiple taps.
Mobility Services: In our evaluation of Bolt’s ride loyalty program, we discovered that unclear thresholds for reaching Gold status discouraged mid-frequency users from continued engagement—despite a high Net Promoter Score overall.
Can AI help us analyze behavioral logs or cluster feedback sentiment at scale? Absolutely—and we use it for that. But will it recognize that the phrase “Not worth it” in user interviews actually stems from unclear point valuation plus a poorly timed email? No. That interpretation requires ethnographic thinking, contextual probing, and pattern-sensing that goes beyond text prediction.
AI Is a Tool. We’re the Architects of the Inquiry.
The best researchers aren’t threatened by AI—they’re using it to clear the noise. At FinUXlab, we incorporate LLMs to summarize transcripts, generate hypothesis frameworks, and rapidly synthesize secondary sources. This allows us to spend more time on what truly moves the needle: framing the right questions, designing methodologically sound studies, and turning observations into actionable strategies.
For example, in a recent diary study of a multinational grocery chain’s loyalty app, AI helped us surface patterns in entries. But it was our human team that noticed a troubling behavioral loop: users were stockpiling points for a large reward that kept getting pushed out of reach due to dynamic pricing—an issue that AI summaries alone never flagged.
What’s Next: Loyalty Programs Need Deeper Research, Not Faster Outputs
As loyalty ecosystems get more complex—integrating payment platforms, cross-brand collaborations, and real-time personalization—our research demands go up, not down. This complexity multiplies the number of assumptions product teams make about user behavior. It also raises the stakes: poor UX in loyalty programs doesn’t just frustrate users—it erodes trust, retention, and ultimately, revenue.
To navigate this, brands need researchers who can:
- Uncover motivations behind point accrual and redemption behavior.
- Design longitudinal studies that track loyalty perception over time.
- Expose emotional friction that quantitative data misses.
- Validate AI-generated insights with real human reactions.
In short: they need teams like ours.
Final Thoughts: Don’t Automate Understanding—Deepen It
AI may be the fastest intern you’ve ever had. But it’s not your strategist, your listener, or your empathetic mirror to real user needs. As loyalty programs become a core lever of brand strategy, companies can’t afford to treat UX research as an afterthought—or an automatable task.
At FinUXlab, we believe the future of loyalty design belongs to those who ask better questions, challenge assumptions, and advocate for user truths in every product decision. And that work, thankfully, still requires a deeply human touch.
