A national coffee chain had a loyalty program with over 600,000 members and years of transaction history. On paper, they had more than enough data. In practice, they weren’t getting much value from it.

Their analytics team could report what customers were doing—visit frequency, spend patterns, seasonal spikes—but the business still couldn’t answer the more important question:

Why were customers behaving the way they were?

Despite regular promotions and a polished app experience, retention had plateaued at 41%, and only 18% of members were redeeming rewards consistently.

What Behavieural Did

Without collecting any new data, we reframed the existing loyalty dataset through a human‑decision lens. We looked for psychological patterns hidden inside the numbers, such as:

• Early‑stage friction causing 27% of new members to drop off within 30 days;

• Reward thresholds that felt too distant, reducing motivation;

• Choice overload in the app that suppressed redemption;

• Offer timing that didn’t align with natural habit windows; and

• Habit patterns that shaped morning vs. afternoon purchasing.

These insights weren’t visible in standard dashboards because no one had been asked to look for them.

What Changed

Working entirely with the client’s existing data and systems, we redesigned:

• The reward cadence, making progress feel achievable;

• The offer timing, aligning with peak habit moments;

• The app flow, simplifying decisions and reducing cognitive load; and

• The messaging, shifting from transactional to identity‑reinforcing (“your usual,” “your routine,” “your spot”).

No new data collection. No new tech. Just better use of what was already there.

The Impact

Within three months, the results were clear:

• Retention increased from 41% to 56%;

• Repeat visits grew by 22%, driven by dormant members returning;

• Reward redemption rose from 18% to 31%; and

• Customer lifetime value increased by 14%.

The client didn’t need more data—they needed a clearer understanding of the people behind it.