Retail is entering a new era of reinvention. At the heart of this transformation is a new breed of AI technology: Large Language Models (LLMs), which are reshaping how consumers discover, engage with, and purchase products. So, what does that mean for retail and retail media?
With ChatGPT now ranked among the top five most visited websites globally, overtaking X (formerly Twitter), and Google’s search dominance dipping below 90% of market share for the first time in decades, it’s clear that the shift from traditional keyword search to dynamic, dialogue-driven discovery is well underway.
And this is only just the beginning. Search is evolving from typing in product names, to asking for nuanced, context-rich help. ‘Find me a family holiday destination where the flight is under four hours, the hotel has an aquapark, and the weather is warm in October’ is the kind of real-world query that shows just how personal and specific consumer queries are becoming.
The launch of Google’s ‘Just Ask Google’ campaign, featuring a character who is reliant on Google in all of his personal every-day decision-making, demonstrates Google’s intent to reposition itself in this new dialogue-driven landscape. In these AI-led interactions, shoppable moments emerge organically from conversation, much like with a real-life sales assistant, opening the door for potential future integration with voice and wearable tech.
For consumers, this means a leap towards contextually aware, hyper-personalised experiences. Imagine an LLM that knows your preferred grocery store, your favourite shampoo brand (and the one you switch to when budgets are tight), or even that you might need nappies based on your history of searching for ovulation kits. This rich combination of data sources from publishers, retailers and user input could become the most powerful live dataset in commerce.
What this means for advertisers: marketing to humans and bots
For advertisers, this represents both a strategic pivot and a significant new opportunity to attract new consumers to both their brand and to their category. As the traditional marketing funnel collapses, and with it, long-held assumptions about how to influence purchasing decisions, advertisers must now consider how they compete not just for human attention, but to the bots that shape their decisions.
This is where new measurement standards will come into play. What is your share of voice within an LLM? How does your content perform in AI- generated conversations to help drive both brand saliency and conversion? Those who start shaping these metrics early will be best positioned to win, both in the eyes of the consumer and the algorithm.
Retailers: the lure of rich data ecosystems
Retailers are poised in prime position to benefit from these changes. Those with large ecosystems, logged-in users and sophisticated data infrastructures can selectively share valuable data with LLMs to power smarter interactions, while safeguarding key datasets for monetisation within their own Retail Media Networks (RMNs).
Retailers such as Kroger, Walmart, Carrefour and Tesco are already demonstrating the power of sticky ecosystems, built on intelligent loyalty schemes and closed-loop measurement. In these cases, collaboration between major retailers and AI platforms feels not only inevitable, but also mutually beneficial.
However, what would happen if some of the bigger retailers chose to withhold certain datasets? For example, if Tesco products didn’t appear in an LLM’s outputs, would that actually deter shoppers from visiting Tesco? Probably not, meaning a retailer could leverage its market power to negotiate on its own terms. This means we could see power struggles similar to what we’ve seen with Amazon blocking other LLMs to their website.
For smaller retailers, the dynamic is far less negotiable. Without robust datasets or significant brand pull, they may have little choice but to share their data to ensure visibility within LLM-powered journeys. In practice, this could mean commercialising access to their information or partnering with platforms like Amazon simply to remain part of the consumer consideration set.
Publishers: content still reigns
Publishers will continue to feel the impact as affiliate revenue continues to be hit by slowly reducing pageviews.
However, content will continue to be king in the LLM world. For publishers, the future lies in creating AI-native content, designed for training and shaping LLM outputs. Success will depend on machine-optimised metadata, structured content, and licensing strategies. We’ve already seen the New York Times licensing its archive to AI companies, which is undoubtedly a smart move to maintain relevance.
The rise of RMNs in an AI era
In this evolving landscape, RMNs stand out as uniquely positioned to play a critical role in LLM-driven search, connecting first-party data, media activation, and closed-loop measurement.
As discovery becomes increasingly dialogue-led and data-driven, retailers that successfully integrate their RMNs with LLM systems will be best positioned for deeper shopper engagement and greater influence over the moments that drive purchase decisions.
AI is no longer just enhancing the shopping experience; it’s reinventing it. Dialogue-driven commerce will replace traditional search. Bots will become gatekeepers. And the fusion of data, content, and intent will define the winners in this latest evolution of retail.
This article was originally published on InternetRetailing.