Asia Pacific has long been a global retail powerhouse, currently home to approximately 60 percent of the world’s consumers. With this scale comes widespread fragmentation, driven by varying levels of digital maturity and technology adoption across the region. Despite this, APAC is leading the world in AI adoption, making it an ideal testing ground for new and emerging technologies, including AI-assisted commerce, where shopping is moving from keyword-based search toward conversational, AI-driven interactions that guide discovery and decision-making.
AI-assisted commerce is compressing discovery
In the traditional e-commerce era, a search for “running shoes” might surface dozens of pages of results. Today, a conversational AI assistant can evaluate the user’s specific intent and return three, five, or even just one recommendation. If a brand does not appear within this curated conversational slot, it ceases to exist in that moment of consideration.
Winning in this environment requires bringing together multiple data layers – past customer queries, recent shopper behaviour, and real-time signals – to surface the most valuable and relevant recommendations at the right moment. We are already seeing the power of this data synchronisation. In Indonesia, H&M partnered with Criteo to combine email lists, past shopper behavior, and lookalike modelling to power AI-based recommendations that reached 12 million unique users and generated 950,000 qualified visits, proving that even in a fragmented market, unified data can drive massive scale.
Retailers at the core of AI-assisted commerce
While large language models (LLMs) provide the linguistic base for conversational retail, they often lack the intricacies of actual commerce, such as localized inventory, real-time pricing, and a deep understanding of shopper behaviour and preferences. This is where retailers play a critical role, as the owners of the transaction and the richest source of commerce signals that power AI-assisted experiences.
Danone’s partnership with Shopee illustrates this in practice. By refining shopping category targeting, competitor store insights, and abandoned-cart segmentation, the campaign reached over 30 million unique users and delivered a 55 percent uplift in order value. It’s a clear example of how retailer transaction signals can be activated to drive stronger marketing performance, and increasingly, to inform more intelligent, AI-driven discovery and decision-making.
As these experiences evolve, consumer concerns around privacy will inevitably grow. The presence of retailers in the equation enables closer enforcement of responsible data usage, without compromising the relevance of recommendations. In markets like Singapore and South Korea, where expectations around data transparency are particularly high, this retailer-led accountability will be a meaningful differentiator.
In short, AI-assisted commerce isn’t here to replace retailers; it’s here to amplify their role.
Product data becomes the new media strategy
AI-assisted commerce fundamentally changes how brands compete for visibility. Rather than ranking results by bid price, an AI assistant evaluates structured inputs – product completeness, price competitiveness, ratings, availability, and more – to decide what to recommend. In this new paradigm, a brand does not outbid a competitor; it out-describes them. Thus, high-quality, structured product data is no longer a back-end IT requirement, but the core of the modern retail strategy.
Retail media becomes a profit engine, not just an ad channel
Most retailers today have their merchandising teams, media sales teams and experience teams operating in silos, each optimizing for a different metric, often at the expense of others. The agentic AI era demands harmony. The most effective retail media strategies today are now full-funnel and cross-channel, combining on-site formats that capture shoppers at the point of purchase with off-site campaigns that re-engage them across the broader internet.
Uber Advertising’s partnership with Criteo in Australia is a compelling example of what on-site retail media can achieve. By deploying Sponsored Products and Commerce Display formats across the Uber and Uber Eats ecosystem, brands across categories like snacking, confectionery, and alcohol reached a high-intent, convenience-driven audience at the moment of purchase. A snacking and confectionery campaign drove a 15 percent revenue share increase in its core category, with a halo effect lifting PDP (Product Detail Page) share in adjacent categories by 11 percent. An alcohol campaign saw even stronger results – a 54 percent surge in revenue share. These weren’t just visibility plays; they are measurable shifts in market share, driven by always-on, full-funnel engagement within the app ecosystem.
Zepto’s partnership with Criteo in India completes the picture on the off-site side. Rather than treating retail media as a standalone channel, Zepto extended its media offering across the open internet, using AI-powered, personalized campaigns to re-engage shoppers and drive them back to convert. In a quick commerce sector growing at 75 to 100 percent year-over-year, getting this integration right early is what separates a sustainable retail media business from a short-term monetisation play.
The road ahead
AI-assisted commerce is introducing a new level of precision to the digital economy. Shoppers are regaining their time through compressed discovery; retailers are solidifying their role as the owners of the transactions, and brands are finding a more direct, data-driven path to their customers. Those that thrive will be those that embrace this shift, prioritizing data integrity over bidding budgets and moving from being just another search result to being the specific recommendation that earns a consumer’s trust.
Szi-Wei Lo is the Executive Managing Director, APAC, Criteo. He is experienced in driving business growth and cultivating high-performing, customer-centric teams across the APJ and Middle East markets, delivering tailored strategies that resonate across diverse cultural and economic landscapes.
TNGlobal INSIDER publishes contributions relevant to entrepreneurship and innovation. You may submit your own original or published contributions subject to editorial discretion.
Featured image: TungArt7 on Pixabay
How DeepSeek changed the rules of Baidu SEO – what international brands need to know in 2026