Has Google just made your loyalty program invisible?

In January 2026, Google published the Universal Commerce Protocol, an open standard for agentic commerce currently in early rollout, that allows AI agents to shop on a customer's behalf, querying merchant systems in real time, comparing prices, applying promotions and completing purchases without a human visiting a website or opening an app. For loyalty programs, the question is whether your benefits will be visible when that agent makes the decision.

Most loyalty leaders are treating this like a technology issue. The harder problem is visibility: can an agent actually see the benefits at the point of decision. If your program's benefits aren't structured for an AI agent to find, read and apply in real time, they don't exist at the moment the decision is made.

The agent moves on. Your brand doesn't get chosen. And nobody flags it as a loyalty failure, because technically, the sale just went somewhere else.

If even a fraction of transactions shift to agent-mediated decisions over the next 12 months, most loyalty programs will see declining redemption and unexplained share loss, with no change to the program itself.

Don't miss any news, updates or insider tips from CX Network by getting them delivered to your inbox. Sign up to our newsletter and join our community of experts. 

Learn More

Your loyalty program has a new customer, and it's already shopping

Australian loyalty program owners currently rate their AI preparedness at just 6.2 out of 10, among the lowest of any English-speaking market surveyed globally, according to Antavo's 2026 Global Customer Loyalty Report. That's the starting position for a market where agentic commerce is already live and accelerating.

The new customer isn't browsing. It's querying. It doesn't respond to creative, doesn't notice your brand values and has no patience for a system that makes it work to find what's on offer. It simply moves to the brand whose program it can read.

If your benefits can't be read in real time, do they exist?

As UCP expands to include Identity Linking for loyalty programs, brands that haven't structured their benefits for machine-readable access will find themselves locked out.

Google's implementation documentation confirms that loyalty recognition is designed to require a separate Identity Linking integration, an optional step on top of the core protocol. Without it, agents complete guest checkouts with no loyalty recognition at all.

The brands building that infrastructure now will be the ones the agent recommends when it becomes the default shopping experience.

Talon.One's 2026 Loyalty Strategies report identifies agent-readable incentives as a core requirement, loyalty benefits must be structured, real-time and machine-readable for an AI agent to act on them. That preparation matters more than most loyalty leaders currently recognize. Antavo's research found 92.7 percent of program owners report positive ROI at an average return of 5.3x, with existing program owners dedicating over half of their marketing budgets to loyalty and CRM. The question is whether an AI agent can find that investment at the moment the decision is being made.

This is a CX problem, not an IT problem

Most loyalty programs were built for a customer who remembers to scan their card, checks their points balance in an app and responds to an eDM.

AI agents solve the program-overload problem that has been building for years, 34 percent of people say they belong to too many programs to keep track of, according to the ALA's 2025 Loyalty Report. Agents track everything, forget nothing and optimize every transaction automatically. The brands structured for agent-readability get chosen. The ones that aren't get optimized around.

The real question isn't whether your tech team has implemented UCP. It's whether the brand proposition you've invested in, your loyalty offer, your personalization strategy, your member value, can survive in a world where the customer's AI agent is making the decision, not the customer.

The consequences are already playing out locally. At the start of 2026, Woolworths upgraded its AI assistant Olive with Google's Gemini platform, making it capable of applying loyalty discounts and placing items into a customer's basket. The same month, Olive made national headlines for telling customers about its mother. Woolworths' response was instructive: they built eight proprietary "agentic judges", background checks that vet every response for accuracy, compliance, and mission completion before Olive surfaces anything to a customer.

The technology had moved faster than the human judgment needed to govern it. So they built the governance layer. That's what designing for agentic AI deliberately actually looks like.

In almost every loyalty audit I run, what a brand thinks its program delivers and what a customer can actually access, let alone an AI agent, are different documents. In a retail program I audited recently, members could only recall around 32 percent of the program's benefits. If humans are missing most of what's on offer, an unstructured system will surface even less.

The loyalty visibility test

This doesn't require rebuilding your program from scratch. It requires passing a test most programs haven't been designed for. I call it the Loyalty Visibility Test. Three layers, in order:

1. Visible: Can an AI agent find your program benefits at all? Not in your app or your terms and conditions, but exposed in real time via structured data at the point where the agent is making the decision.
2. Verifiable: Can the agent confirm a customer's eligibility, tier status and entitlements instantly? If it has to wait for a batch update or query a siloed CRM, it won't wait. It will move on.
3. Actionable: Can it apply those benefits automatically at checkout, points redemption, member pricing, exclusive offers, without a human logging in to trigger it?

Right now, most loyalty programs can't pass any of these three. The infrastructure simply isn't there yet. But the brands that build toward all three now will be the ones the agent recommends when it becomes the default. The ones that don't will just stop being found.

The map most loyalty programs don't have

The starting point isn't a technology audit. It's a customer journey audit. Most brands haven't mapped where AI has already entered their customer experience, the algorithmic discovery moments, the automated service loops, the agentic touchpoints that now sit alongside every human interaction.

Until you know which moments an agent is influencing, you can't know which parts of your program are visible, which are invisible, and which represent the greatest commercial risk.

That map needs to account for more than agent-readability. An agent can apply your points. It can't decide that this particular member deserves something more. That's a decision a brand has to make deliberately, before the agent is involved. The ACTIVATE Framework™, an operating model for AI-era CX and loyalty design, gives loyalty leaders a structured way to audit the real journey, close the visibility gap and protect the human moments that automation can't replace. The full framework is available here. 

The question worth asking now

This isn't a simple human-versus-AI debate. The real question is whether your program can be found, read and acted on by the entity now making purchasing decisions on your customer's behalf, and whether the moments that actually create loyalty are being designed deliberately or left to chance.

Agentic commerce won't announce itself as a loyalty problem. It will just show up in the numbers, slowly, until a CFO asks why redemption is down and nobody in the room has a good answer. The brands that close that gap first won't just protect their loyalty ROI. They'll be the ones the agent recommends.

Quick links

• The persona problem: What agentic AI exposes about our favorite CX tool
• When your customer is a machine: Rethinking service design for AI agents
• How generative AI search is changing the customer journey

CX HORIZONS: The state of CX in 2026

Download our report to learn the results of CX Network's research into the top CX trends,  challenges, spending plans, and customer behaviors driving CX in 2026. Hear from thought leaders on what you should be doing now to ensure your organization is on track to become a strategic growth engine.

Download Now