Challenge

LPP had no shortage of tools or reports. But despite a wealth of dashboards and internal analytics, the team struggled to turn data into direction. They needed to understand not just where people clicked, but why. They wanted to detect friction before it impacted conversion and understand the real-world behaviors shaping customer experiences.

At the same time, launching and scaling mobile analytics was far from simple. With hybrid code, a wide range of devices, and no standardized screen structure, their team couldn’t rely on out-of-the-box solutions.

Goals

• Decode complex mobile user behavior: With 186 unique views inside the app, our client needed a clear way to track, organize, and understand how users navigated across sessions.
• Connect loyalty actions with business outcomes: The app was largely built around loyalty - rewards, point redemptions, and member experiences. But it was unclear how these behaviors tied to actual purchases.
• Spot friction before users speak up: Rather than wait for negative feedback or declining metrics, the team wanted to catch signs of confusion and frustration as they happened.
• Establish a scalable mobile analytics process: Unlike web platforms, mobile analytics required tailored tracking and new ways of reading behavioral data. The goal was to make this process smooth and repeatable.

Our approach

1. Building on early momentum

After a successful web pilot that mapped the online purchase journey, LPP decided to move forward and work with CUX on mobile analytics. The mobile app was becoming their key channel and a central piece of their strategy to compete with global retail giants.

2. Setting up mobile analytics in a fragmented ecosystem

We kicked off a pilot project to implement full behavioral analytics in the mobile app. While web integration typically takes under 10 minutes, mobile was a different story. It required deep collaboration between development teams and multiple iterations.

In total, the integration took six working days. We provided the documentation and SDK, and the client’s team released an SDK-ready version of the app for testing. During the production release, we worked with LPP's developers to identify and eliminate app crashes and SDK conflicts. We discovered that aggressive code compression was blocking autocapture for clicks. Our teams collaborated to find a solution, which was implemented in the next app release. After six days of iterative work, click tracking and behavioral capture were fully enabled.

3. Mapping user journeys across 186 unique views

Unlike the web, where journeys follow predictable URLs, the mobile app contained 186 different views. For non-developers, it was nearly impossible to know which screen meant what. To make sense of this complexity, we delivered exploratory tools that revealed previously invisible user actions:

• Auto-generated event lists
• Screen registration and naming
• AI-powered path visualization

These tools gave the client’s team a structured way to explore data without needing to decode technical screen names or manually reconstruct user flows.

4. Uncovering key behavioral insights

As soon as the data started flowing, two patterns stood out:

Loyalty-first behavior: Nearly half of all mobile visits were loyalty-driven. Users weren’t just browsing; they were coming with purpose to explore coupons, redeem points, open gamified scratchcards, and check in to earn rewards. This proved the app was attracting an engaged, high-intent audience.

The cart as the center of intent: The shopping cart was the key marker distinguishing casual visits from purchase-oriented sessions. We found that most users added items from product pages (after seeing photos) and frequently used the cart to review saved items and wishlists. These micro-interactions, often missed by traditional analytics, were strong indicators of purchase intent.

We also discovered that 1 in 3 users who completed a purchase returned to the order confirmation screen. They were looking for reassurance that their transaction was successful. This single insight led to immediate UX improvements, like adding clearer progress indicators and visual confirmation cues.

Impact

Go-live in under a week

Despite technical complexity, behavioral analytics were fully operational within six days thanks to close cross-team collaboration.

Demystified complex mobile journeys

Exploratory tools made mobile data readable, allowing the team to understand user flows without needing technical expertise.

Loyalty strategy reshaped by data

The finding that nearly 50% of sessions were loyalty-focused proved that these users were highly engaged and provided a clear catalyst for conversion.

Fast UX wins based on real behavior

Insights into the cart and post-purchase behavior allowed the team to act quickly on findings that traditional analytics would have missed, improving the user experience immediately.

What our client said

"We care deeply about the experience of our mobile customers, especially those who have decided to download our app. They are our most valuable users in many respects, both from the perspective of conversion and in the long-term view of Customer Lifetime Value (CLV). That's why we want to know as much about them as possible.

It’s no surprise that large e-commerce apps like ours have highly developed analytics, and their usability is also regularly studied by a research team. CUX, however, allowed us to look at the behavior of our mobile app customers from a new perspective - the very one we were looking for to complement the capabilities we already possessed.

Of course, this is a very difficult technological challenge in native applications, which was a concern for us. It requires an innovative approach, but as you can see, we managed to work together to deliver knowledge and discover things we were previously unable to see."

Marcin Wcisło, Head of Experience Design, Silky Coders