Your Loyalty Program Just Moved Into the Chat Window

Imagine this. Your best customer never opens your app this month. Never visits your website. Never sees your email.

And yet they earn points, redeem a reward, and get a surprise birthday treat — all inside a chat window, through a conversation with an AI agent.

This is not a far-off idea. It is already happening.

On September 29, a new AI-powered loyalty platform called Pointics launched for online stores. It lets shoppers earn and redeem points the usual way. But its roadmap has something more interesting: it plans to let AI assistants and agents connect to your loyalty data. Soon you could ask, "Which customers are close to their next VIP tier?" or "Who has points sitting unused?" — and get an answer in plain words, not a spreadsheet.

Meanwhile, back in April, a UK grocery chain teamed up with a major software company to put agentic AI inside its loyalty program. That means the AI does not just answer questions. It takes actions: it can spot the right moment to offer a reward and trigger it.

The loyalty program is leaving the app. It is moving into the conversation.

The chat window is the new front door

People now shop by chatting. They ask an AI assistant which running shoe to buy. They ask it to reorder the dog food. They ask it to compare prices.

A 2025 shopping index from Klaviyo found that 68% of consumers prefer AI for instant answers, 56% prefer AI for personalized recommendations, and 48% prefer AI for post-purchase support. PYMNTS research found that more than 6 in 10 US consumers used AI platforms like ChatGPT, Claude, or Gemini in the past year.

Here is the part that should worry loyalty managers: more than half of AI users would rather complete a purchase inside the AI platform than get sent to a merchant's website.

Think about what that means. If your points, tiers, and rewards only live on your website, they are invisible at the exact moment a customer decides what to buy.

From points on a shelf to rewards that come to you

Loyalty used to be passive. You collected points. They sat on a shelf. Maybe you remembered to use them. Most people did not. Industry research shows only about half of loyalty members actively engage with their programs.

AI agents flip this around. Instead of the customer remembering to check their balance, the agent remembers for them. It notices you have 2,300 points. It notices a reward you qualify for is about to expire. It mentions it while you are chatting about something else entirely.

That is a small change in words but a big change in behavior. Redemption goes from a chore to a pleasant surprise. And redemption matters: when people actually use rewards, they come back. Loyalty members drive 12%–18% more annual revenue than non-members, according to industry statistics. The hard part was never the points. It was getting people to use them.

Agents that notice when a customer is drifting away

The most powerful use of AI in loyalty is not rewards. It is noticing.

Right now, "silent churn" — customers who stop buying without complaining — is one of the hardest problems in retail. By the time you see it in a monthly report, the customer is gone.

An AI agent watching the chat channel can catch the early signs: the customer who used to order weekly now orders monthly. The one who always bought premium now buys the cheapest option. The one whose last three chats sounded frustrated.

With the loyalty data in front of it, the agent can act right there in the conversation — a gentle nudge, a bonus for coming back, a question about what went wrong. Not a generic "we miss you" email three months later. A response in the moment, in the channel the customer is already using.

What changes for your growth team

Three things.

First, your loyalty data must be machine-readable. If your points, tiers, and rewards live in a closed system that only your app can read, no AI agent can work with them. The winners will be brands whose loyalty programs an agent can ask about and act on.

Second, stop measuring loyalty like a filing cabinet. Old metrics: how many people signed up, how many points are outstanding. New metrics: how many rewards were redeemed inside a conversation, how many drifting customers were rescued in the chat window, how much of your revenue started inside an AI chat.

Third, keep the human feel. People like chatting with AI, but they still want it to feel like the brand. The agent should know the customer's name, their tier, their favorite products — and act like someone who has been paying attention, not a vending machine.

The takeaway

The loyalty program of the next five years will not be a card in your wallet or a tab in an app. It will be a conversation. Customers will ask an AI what they should buy, and the AI will quietly fold in their points, their tier, and the reward that is about to expire.

The brands that win will not be the ones with the most generous points. They will be the ones whose loyalty programs show up in the chat window — at the exact moment the decision is made.

Sources: Pointics launch announcement (Sept 2026); Klaviyo; PYMNTS; 99minds; WiserReview.