Evolving eCommerce Loyalty: From Pre-COVID Simplicity to AI-Powered Personalization

Introductory Context

Earlier this week, while diving into one of my regular reads, I came across a thought-provoking article from my alma mater, IESE Business School, about how AI can revolutionize customer loyalty in retail. It struck a chord because loyalty programs have always been a cornerstone of my work in product leadership.

As I reflected on how loyalty systems were designed before COVID—and how they would evolve in today’s AI-driven world—it became clear how much has changed, yet how much has stayed the same. While technology and expectations have advanced, the basics of great product design remain the foundation of building impactful customer experiences.

Let’s take a look at how loyalty programs were built in a pre-COVID world, how they might be built today with AI, and why the fundamentals of product design are still at the heart of it all.

Building Loyalty Programs Pre-COVID

Pre-COVID, loyalty programs were often built with simplicity and scalability in mind. I remember leading the creation of a multi-brand loyalty system for 160+ bus operators, where the primary goal was to unify various reward structures into a single, centralized framework.

Back then, the system relied on:

• Static reward tiers: Customers earned fixed points based on ticket bookings, with predefined milestones for rewards.
• Transactional engagement: The focus was on driving repeat purchases rather than building deeper emotional connections.
• Static data insights: Data was primarily historical, offering limited predictive power for retention or engagement.

While this approach was effective at the time, it came with challenges. The lack of personalization made it harder to retain customers who wanted more tailored experiences, and scaling the system, say with other OTA partners, required API integration effort.

How Post-COVID AI Would Transform the Same Product

Fast forward to today, and the way we would approach building a loyalty system is vastly different. With AI in the mix, the same product would focus on personalization, automation, and predictive capabilities.

Here’s how:

• Dynamic rewards: Instead of static tiers, AI would adjust rewards based on customer preferences, behaviours, and predicted lifetime value, ensuring every interaction feels personal.
• Proactive engagement: Predictive analytics would flag customers at risk of churn, enabling timely interventions like exclusive offers or personalized content.
• Seamless partner integration: Automated APIs would streamline onboarding, drastically reducing the time needed to scale the program.
• Real-time insights: AI would analyse customer behaviour in real time, allowing the program to adapt dynamically and stay relevant.

This shift not only enhances the customer experience but also significantly improves operational efficiency. With AI handling much of the heavy lifting, teams can focus on strategic improvements rather than manual processes.

A Comparison of Then vs. Now

Looking back, the pre-COVID approach had its strengths—it was straightforward, scalable (to an extent), and met the needs of a less competitive market. But it lacked the personalization and adaptability that today’s customers demand.

In contrast, the post-COVID approach harnesses the power of AI to create dynamic, personalized experiences that engage customers on an emotional level. By predicting churn, adapting rewards in real time, and automating processes, the program becomes not just a transactional tool but a relationship-building platform.

That said, the fundamentals of product design tie these two worlds together. In both cases, success hinges on:

• Customer understanding: Whether it’s designing static tiers or AI-driven rewards, the starting point is always understanding what customers value.
• Iterative development: Both approaches benefit from testing, learning, and refining to align with customer needs.
• Cross-functional collaboration: Bringing together data scientists, engineers, and marketers ensures that the product is not just innovative but also practical and impactful.

Pros and Cons of Each Approach

Pre-COVID:

• Pros: Simplicity, quicker to develop, less dependency on advanced technology.
• Cons: Limited personalization, labor-intensive scaling, reactive rather than proactive.

Post-COVID (AI-Powered):

• Pros: Highly personalized, efficient scaling, proactive retention strategies.
• Cons: Requires significant upfront investment in AI infrastructure and data privacy safeguards.

While the post-COVID approach offers transformative potential, it’s critical not to lose sight of the fundamentals. Technology alone doesn’t solve problems—strong product design ensures that innovations truly meet customer needs.

Why Product Design Always Wins

No matter how advanced the technology, the basics of product design remain constant:

1. Start with the customer: Deeply understand their pain points and design solutions around them.
2. Simplify complexity: Even the most sophisticated AI systems should feel seamless and intuitive for users.
3. Iterate relentlessly: Build, test, learn, and improve. The best products are born from continuous refinement.

These principles have guided my work across industries, from unifying multi-brand loyalty systems pre-COVID to envisioning AI-powered solutions post-COVID. They’re the bridge between past and future success.

Conclusion

It’s moments like these—connecting the dots between thought-provoking research and real-world applications—that remind me why I love what I do. Staying curious and committed to great design will always be the foundation for meaningful innovation.

Let’s connect and brainstorm how we can shape the future of eCommerce! Let me know your thoughts—how have you adapted your product strategies to address these shifts? I’d love to hear your experiences!