Loyalty as the coordination layer across the whole customer relationship
By Adriana Sofia Gomez · Director, Strategic Consulting
The best loyalty teams are no longer built as marketing sub-functions. They are built as a coordination layer across the entire customer relationship: recognizing a customer, deciding the next appropriate action, and directing the response across every team that engages with the customer. That shift changes what a loyalty team is for, how it’s structured, and the capabilities it must own.
When a brand is building or evaluating its loyalty team, we encourage leaders to think in terms of capabilities rather than job titles. Smaller teams often cover several capabilities with one person, and that works well. The pattern that causes trouble is when one of the four key capabilities has no owner at all.
The Four Capabilities Every Loyalty Team Needs
Orchestration authority.
One accountable leader decides what the brand does for a given customer in a given moment: which message a person sees, which channel carries it, and what offer (if any) is needed. That role needs to sit at the same level as the functions it coordinates, e.g., reporting to a CMO, chief customer officer, or business-unit leader, with its own budget and a formal voice in experience design. When positioned a layer or two down, loyalty ends up requesting cooperation from CRM, digital, and frontline teams rather than directing it.
The data and decisioning spine.
Data includes identity resolution (recognizing the same person across the app, the website, the store, and the contact center) and zero-party data (what members tell you directly about themselves). The strongest programs build toward what Kobie calls a Panoramic Customer Profile™, stitching together transactional, behavioral, emotional, and always-on zero-party data capture over time rather than in one exhausting ask.
Decisioning is the logic that acts on that data: whether it is centralized, whether it learns from what worked the last time, and whether it uses signals while they’re still fresh.
AI raises the stakes for the data and decisioning component of loyalty. Data strategy and loyalty strategy have become inseparable, because AI amplifies whatever foundation it runs on. Build AI-driven personalization on shaky, slow, or thin data and it won’t deliver the returns you’re expecting; it will scale the same irrelevance, faster. In our experience, even the most advanced models can’t produce reliable outcomes without the high-fidelity data that loyalty programs are uniquely positioned to capture.
Program economics.
This capability covers incrementality measurement (would the purchase have happened without the incentive?), liability and total-cost management, and visibility into performance by channel. When it has no owner, programs optimize for redemption rate instead of incremental customer value.
Owning the loyalty program well also means bringing finance into a shared view of its value: a program treated as a marketing cost line gets managed for spend, while one that finance recognizes as a driver of incremental value earns a different level of investment.
Experience design.
For an enrolled member, the program becomes the most consistent expression of the brand they encounter. The welcome, the points balance, the birthday, the service recovery, the redemption moment – all happen within the program. Someone should own that whole arc, including how it shows up in CRM, in the app, in person, and in the sales conversation. Left unowned, loyalty becomes a set of mechanics hiding behind other channels instead of a unified brand expression.
How Ownership Works Across the Four
The brands that handle cross-functional loyalty well tend to separate ownership of the customer profile from ownership of day-to-day execution. Loyalty owns the identity and decisioning described above, which gives it a legitimate basis to advise other functions on experience, delivery, and measurement. Marketing, technology, finance, and operations own delivery within their own channels, all working from the same view of the customer.
Executive sponsorship is the connective tissue that holds the arrangement together; without it, loyalty remains a program that other functions can quietly opt out of.
What’s Driving the Reshuffle
Two forces have done most of the reshaping over the past few years.
The first is data abundance without a matching increase in decision capability. Industry research finds that close to 60% of enterprise data goes unused in decision making and that only about 15% of brands can deliver personalization in real time. The constraint isn’t the volume of data; it’s the speed at which logic responds to new data.
The second is AI moving from experiment into everyday practice. Kobie’s research found that 74% of practitioners use AI at least weekly, though most of that use is still internal efficiency rather than member-facing experience. Looking ahead, 68% expect their organizations to expand AI use over the next one to two years, with the largest anticipated growth in dynamic rewards, journey orchestration, and real-time personalization.
Teams are responding to both forces. Data strategy, identity, and measurement roles that once lived in IT are moving into the loyalty function, and many teams are staffing ahead of the AI expansion rather than waiting for it to arrive.
The Skills Becoming Harder to Find
As programs become more automated, the scarce expertise shifts from running campaigns to exercising judgment which means recognizing a customer quickly and reliably, and measuring whether an incentive changed behavior rather than paying for a purchase that would have happened anyway.
More of what a program does will be decided by a model than by a rule someone wrote, and the skill is knowing when to hand a decision over, how to tell whether the model is outperforming the rule it replaced, and who answers for it if something goes wrong.
Part of that same accountability is disclosure (being straight with customers about when AI is shaping what they see). Most expect it and can’t detect it on their own, which makes it something a program must own deliberately rather than assume.
Where to Start When Resources Are Tight
Start with a diagnosis, not a hiring plan. Scoring the team against the four capabilities usually reveals that performance is being held back by one weak area dragging on the others, not by underinvestment across the board.
From there, the question isn’t really “build or buy.” It’s how close a capability should sit to the brand’s own context. Program strategy, customer insights, and ownership of the economics draw heavily on institutional knowledge, so they tend to work best as a partnership: the brand supplies context and owns the direction, while an outside partner brings expertise and pattern recognition from across the industry. Platform operations, campaign production, and program analytics can be handed off more completely.
Within the program, prioritize quick wins. Consider a note on a member’s birthday, a thank-you when someone reaches five years with the brand, or a small reward that arrives unannounced for a loyal customer. Gestures like these acknowledge a customer as a person rather than as a source of purchases and run on data brands already have. Knowing when someone joined and when they were born is enough to act on, with no platform purchase, no new data initiative, and no added headcount. These recognition moments can be built once and left to run.
The Loyalty Function of the Future
Over the next few years, we expect the loyalty function to look less like a program team and more like the coordinating layer for customer decisions across the business. The analogy we find most useful is a nervous system: it senses signals (customer data), processes them (decisioning), coordinates a response (orchestration), learns from the result (intelligence), and connects everything (delivery). Loyalty already holds most of those components, which is why the stack keeps reassembling around it.
Measurement will evolve to match. Program metrics like enrollment and redemption rate give way to system-level measures: how many actionable data points exist per customer (data efficacy), what share of customer decisions are made with loyalty context available (decision quality), how many of those decisions produce incremental behavior (decision incrementality), and how quickly the system moves from signal to decision (data velocity). Teams managing a few thousand journeys today will become accountable for orders of magnitude more signal.
Execution will automate faster than judgment will. The skill most worth building now is the ability to frame the right question about a customer and recognize which answer is worth acting on.
The practical starting points are unglamorous and take longer than most leaders expect: build the identity spine, consolidate decisioning logic in one place with a feedback loop attached, and establish measurement discipline before the volume scales.
Explore more insights from the Kobie team at kobie.com/insights.
