How AI is reshaping what it means to convert in online retail

A year ago, 38% of consumers used generative AI somewhere in their online shopping journey. Today it is 51%. Traffic from AI sources to US retail sites has increased by 4,700% year over year. The typical reaction to numbers like these is to call it a trend and wait for more evidence. The problem is, the brands waiting for more evidence are already behind.

AI is transforming online retail in three concrete ways: how products are discovered, how purchase decisions are made, and how brands keep customers coming back. Each of these shifts has specific implications for how you should think about conversion, loyalty, and the digital experiences you build.

At Livewall, we design and build digital experiences for consumer brands across retail, FMCG, and entertainment. We work at the intersection of these three shifts, and the patterns we are seeing across client work are consistent with what the broader data shows.

How AI is changing product discovery

The first shift is in how people find products. AI-powered search does not just match keywords. It understands intent by combining behavioral context, browsing history, and preference signals. A shopper asking an AI for "a gift for someone who enjoys slow mornings" gets different results depending on that person's prior purchase history, even if the words are identical.

For brands, this changes what it means to be findable. Keyword optimisation gets you into a search index. Behavioral signal optimisation gets you into AI recommendations. The brands that win the second game are the ones whose customers have expressed clear, persistent preferences that the AI can act on.

This is where ongoing loyalty mechanics matter more than ever. Every interaction a loyal customer has with your brand, every product rated, every preference expressed, is a signal that feeds into how visible your brand becomes when an AI is deciding what to surface.

We have seen this with the Decathlon game loyalty activation, which tied member data to real sport preferences. Members who engaged consistently provided richer behavioral signals, reinforcing brand relevance well beyond the campaign itself.

The conversational commerce shift

The second shift is in the consideration layer. 73% of consumers now use AI somewhere in their shopping journey. The most common use cases are product discovery (45%), review summarisation (37%), and price comparison (32%).

The practical implication is that consumers are arriving at purchase decisions pre-qualified by AI. By the time they reach a product page, or ask their AI agent to complete the purchase, the consideration phase has largely already happened inside a conversation.

AI-assisted chat converts at 12.3% versus 3.1% for unassisted browsing. That 4x difference reflects the pre-qualification effect. A consumer who has used AI to guide their purchase is more confident and more committed before they reach the decision point.

Brands that build AI-assisted shopping into their own platforms capture this effect for themselves. Those that don't are ceding the entire consideration phase to whatever AI agent the consumer happens to be using, with no control over how their brand is presented or filtered.

What this means for loyalty in an AI-mediated world

The third shift is in how loyalty works. When AI mediates shopping, transactional loyalty, buying from the same place because it is familiar, becomes harder to maintain. A consumer who delegates purchasing to an AI agent will not necessarily end up at the same retailer they used last week.

What survives is behaviorally grounded loyalty. The kind built on genuine brand attachment, real preference expression, and direct digital relationships that do not run through an intermediary.

Consumer brands that have invested in gamified loyalty and owned engagement mechanics are better positioned for this shift. They have behavioral data that AI can work with. Their customers have expressed preferences actively, not just passively shopped.

Personalisation only works when you have data worth personalising from. And you only get that data through direct, meaningful engagement with your customer base. The HEMA Stapelgek gamified loyalty programme is a good example: everyday purchases became a game mechanic inside the app, driving consistent behavioral data alongside consistent engagement.

At Livewall, we help brands build these engagement foundations. The first-party data strategies, the loyalty mechanics, and the owned digital experiences that make personalisation meaningful and loyalty durable in a world increasingly shaped by AI.