Skip to content
Fast Company Executive Board

The Fast Company Executive Board is a private, fee-based network of influential leaders, experts, executives, and entrepreneurs who share their insights with our audience.

Future-proof your loyalty program for an AI world

AI exposes the cracks in loyalty programs

[Photo: Getty Images]

Profile photo of Enrique Muñoz Torres

Cardlytics

Loyalty programs have long been a cornerstone of retail. They help brands retain customers, encourage repeat purchases, and build differentiation in crowded markets. Loyalty programs are also about data: Capturing what shoppers buy, how often, and increasingly, why.

But AI has completely changed the game, and now almost every retailer is racing to layer artificial intelligence into their processes. The data fueling those models is absolutely critical. If loyalty program customer data is riddled with gaps, errors, or blind spots, the shiny new AI will quickly show its cracks.

Here’s the paradox: The very customers most engaged in loyalty programs (i.e., the “best” customers) are likely driving AI insights with detailed profiles, robust purchase patterns, and rich engagement. But that leaves the remaining—typically much larger—shopper base underrepresented in the dataset.

A distorted picture results. The image of your top-tier shopper base looks sharp and vivid. The rest is nowhere as clear. Without a strategy to close this gap, AI risks optimizing against incomplete truths. It’s the classic problem of “garbage in, garbage out,” amplified by AI’s speed and scale.

The gaps are larger than many realize. According to Bluecore’s 2024 Customer Growth Benchmarks Report, nearly 70% of in-store transactions cannot be tied back to a shopper. Data directly from customers at Bridg confirms these percentages.

Feeding those blind spots into AI models wastes marketing spend, distorts demand forecasts, skews inventory decisions, and undermines customer lifetime value predictions.

Two of the most common AI applications in retail today are:

  • Customer service chatbots: These rely on accurate context to resolve issues efficiently. While most chatbots request order details like email, name, or order number, their ability to resolve problems seamlessly depends on the data already available about that customer.
  • Marketing activation tools: These tools optimize in real-time for channel, product, offer, and timing. The tools require consistent and complete data to ensure the right messages reach the right people.

AI tools can miss the mark when driven by inconsistent underlying data (e.g., a shopper uses multiple loyalty cards, an employee mistypes a customer number, or an in-store purchase is made anonymously).

The outcome isn’t always a complete collapse, but rather a subtle erosion in effectiveness, including misaligned offers, wasted impressions, and missed customer-service opportunities.

Loyalty data alone limits. The system is only as strong as the information captured, and programs capture only part of the truth. Loyalty programs, as they exist today, won’t be enough to power an AI-driven retail future.

Retailers need to strengthen and complete their customer dataset to gain a clearer view of customer identity and behavior for better inventory planning, more precise campaign measurement, and AI models that optimize against reality, not noise. This requires a two-pronged, transformative approach.

 Zero-party data strategies

Consumers willingly share zero-party data, typically in exchange for immediate value. The key is reciprocity—shoppers expect something in return for their information.

Gamification and progressive profiling allow retailers to gather richer insights over time, in ways that feel engaging rather than transactional. Real-time, AI-driven personalization shows customers that the more they share, the smarter the program becomes.

Retailers enrich the dataset and address common inactive loyalty program problems by expanding rewards to include referrals, app engagement, or content interactions.

After all, the average U.S. consumer belongs to more than 15 loyalty programs but actively uses fewer than half, according to Boston Consulting Group.

In-store identity resolution

Even with better loyalty data-gathering techniques, information can still fall short. Shoppers use multiple cards, skip enrollment, provide outdated details, or choose not to identify themselves.

In-store transactions, in particular, represent a massive blind spot. Without identity resolution, purchases often remain anonymous, creating significant gaps in the dataset.

By adopting robust identity resolution solutions, retailers bridge this divide, connecting transactions to real customers, eliminating anonymous purchases, and doubling the universe of shoppers they can understand and reach.

Final thoughts

Loyalty programs aren’t going away. In fact, they are more important than ever as retailers seek to build durable connections with their customers. But the future of loyalty lies not in the points or perks, but in the quality and completeness of the data these programs generate.

Preparing for an AI-driven future means ensuring that data is comprehensive, accurate, and actionable. Combining modern zero-party data strategies with in-store identity resolution allows retailers to close the critical gaps loyalty programs leave open.

The payoff will be twofold: AI models optimizing against real customer data, and stronger, longer-lasting relationships with shoppers. In an evolving AI-powered retail world, loyalty without accurate identity resolution isn’t loyalty at all. It’s another blind spot waiting to be exposed.

Enrique Munoz Torres is the general manager of Bridg.