Increasingly, shoppers are bypassing search engines, comparison sites and even traditional e-commerce platforms. Instead, they ask conversational agents like ChatGPT or Perplexity to evaluate options, summarise reviews and identify the best place to buy the item.
Image: Shutterstock
AI tools that act on behalf of consumers are quietly but decisively shifting power in the retail landscape. Increasingly, shoppers are bypassing search engines, comparison sites and even traditional e-commerce platforms. Instead, they ask conversational agents like ChatGPT or Perplexity to evaluate options, summarise reviews and identify the best place to buy the item. Some agents are already completing purchases.
This change alters more than just shopping habits. It challenges the role of traditional intermediaries and redistributes influence between retailers and brands. In this environment, those who control access to the end customer and meet their needs in machine-readable ways will hold the advantage.
These insights summarise research published in our recent Harvard Business Review article that explored how AI agents are transforming consumer search and buying behaviour.
From balance to imbalance
In the early days of retail, brands and retailers shared a relatively equal footing. Retailers tracked receipt-level sales – what was bought, when and in what combination – while brands relied on aggregated “sell-through” data that estimate that quantities that actually reach the end consumer. Because neither side had a full picture, collaboration was key. Brands conducted in-store research, and retailers shared point-of-sale data to improve promotion and supply chain planning.
The arrival of e-commerce changed that balance. Online retailers gained direct access to customer-level data, including browsing patterns, returns and personal preferences. This shift allowed them to optimise pricing, tailor product recommendations and eventually build entire advertising businesses from the data. As a result, retailers became gatekeepers not just of transactions but of attention and influence.
Also read: Aravind Srinivas’ Perplexity AI is changing the way we search for information online
A second power shift begins
The rise of the internet gave ecommerce retailers controlling access to the end customer massive power to extract profits. A new shift is underway. AI agents are making retailers even less central to how consumers search and decide what to buy. These tools can search across hundreds of sources and weigh options against consistent criteria: price, availability, delivery speed, service quality and return policies. Unlike easily overwhelmed humans, they don’t default to familiar sites or skip over smaller players. Their scope is broader and their standards less forgiving.
Not limited to a handful of trusted platforms, AI agents can surface alternatives that might otherwise be missed, such as a boutique designer with better reviews, or a local supplier with faster shipping. In doing so, they level the retail playing field and transfer power to those offering objective value, not just brand awareness or prime digital shelf space.
Retailers like Amazon, with efficient logistics and user-friendly service, are also well positioned. Their ability to meet the high standards of a GenAI agent will cement their dominance. But others, especially mid-tier players that offer neither standout service nor pricing, will face increasing pressure. Differentiation will matter more than ever, and data transparency will make it harder to fake.
Brands face a new visibility test
For brands, this shift creates both risk and opportunity. It will no longer be enough to compete on name recognition. AI agents are designed to detect equivalence. If a generic light bulb is functionally identical to a premium brand’s version, they will recommend the cheaper option. In categories where differences are hard to perceive or articulate, brands may struggle to command a premium unless they offer something demonstrably superior.
To stay competitive, brands will need to invest in what we call AI agent optimisation (AAO). This means ensuring their product strengths – from build quality to performance to design – are clearly measurable, machine-readable and embedded in the sources agents rely on, such as reviews, specifications and third-party comparisons.
AAO will also compel brand marketers to shift from brand-first messaging to need-first positioning. Consumers and their AI agents are less likely to search for a specific brand, and more likely to ask for the best product for a given scenario: “What’s the best carry-on suitcase for frequent flyers?” Agents are more likely to surface brands that articulate their unique value in this context.
The four levers of distinctiveness
In this new environment, four levers will matter most:Price: Competitive pricing or entry-level models that defend against generic substitutesInnovation: Features, materials or performance that are hard to imitateDesign: Aesthetics and usability that resonate with users and appear in feedbackService: Support, warranties or return policies that show up in agent-indexed forums
Take wireless phone chargers. Ikea wins on affordability. Zens competes with better charging speed and more coils. Both can succeed – if their value is clear to the agent doing the comparison.
The coming era of AAO as a complement to SEO
The path forward is not about abandoning brand building. Rather, it’s about building brands in a way that will speak to the new gatekeepers of consumer choice. Brands must assume that the first encounter with their product may not be through a human decision, but a machine’s logic.
To succeed in this AI-driven retail landscape, brands and retailers will need to clearly set themselves apart through distinct offerings and by building strong AAO capabilities. Innovation, thoughtful design and high-quality service will continue to matter – not just to consumers, but to the algorithms that guide their decisions. This will lead to the rise of AAO as a new discipline, akin to search engine optimisation (SEO).
Conversely, brands that offer little more than name recognition, without tangible value or differentiation, may struggle to retain their position. As AI agents increasingly shape the consumer journey, long-term success will hinge on being both discoverable and meaningfully distinct.
This is an adaptation of an article published in the Harvard Business Review.
]]>
[This article is republished courtesy of INSEAD Knowledge, the portal to the latest business insights and views of The Business School of the World. Copyright INSEAD 2024]
