Your Customer Relationship Is Your Strongest Asset. Protect It.
Last week I pondered, who will own the customer when AI does the shopping? With agentic commerce creating a new coordination layer between retailers and their customers, with technology and payments companies building the infrastructure to power it. The next question is what retailers can do about it, starting with what they already have.
Australian retailers have invested heavily in direct customer relationships for years. Sixty per cent of Australian retailers, and 89% of those with turnover above $100 million, say loyalty and membership programmes are their most effective tool for driving repeat purchases and retention. These programmes were built to compete against other retailers, to create compelling reasons to engage for customers and in exchange, a valuable data asset for the retailer.
Agentic commerce is both shifting and adding to this competitive landscape. The technology and payments companies assembling the new coordination layer are now stepping into the competition for customer relationships and “ownership”.
What Happens When You Lose the Relationship
When a coordination layer forms between a business and its customer, the business controlling that layer captures a disproportionate share of the value.
The Amazon Marketplace is a good example. Third-party sellers now account for 61% of units sold. Research by Pattern and Profitero found that 96.4% of brands selling on Amazon experience price erosion, while 83.3% cite a lack of data or tools to forecast their own sales. Brands face fees averaging 15 to 25%, complete loss of pricing authority and no direct access to the customer. The exchange was distribution for relationship, and the platform progressively captured the value that brands had spent years building.
AI agents represent the next coordination layer forming in retail. They sit between the consumer and the retailer, mediating discovery, comparison and increasingly, purchase. The pattern is recognisable. Retail should take note.
Why the Customer Relationship Is the Moat
The evidence that direct customer relationships create measurable competitive advantage is strong, and that advantage becomes more valuable in an agent-mediated world.
First-party data consistently outperforms intermediated alternatives. BCG benchmarks show that companies using first-party-data-driven personalisation achieve 1.5 to 2.9 times higher revenue uplift than their peers, with more than 25% lower customer acquisition costs. In an agent-mediated world, first-party data is what enables a retailer to power its own AI recommendations, personalise its own channels and understand its customers in ways that no third-party agent can replicate.
Trust compounds this advantage. Research published in the Journal of Retailing and Consumer Services found that trust in a brand’s AI systems is a measurable component of overall brand equity and a critical mediator of purchase intention, suggesting that trust built through years of direct relationships can transfer to AI agents operated by that brand. In Australia, Salesforce found that 69% of consumers say advances in AI make trust more important, and 79% want to know when they are interacting with an AI agent rather than a person.
The moat is real. The question is what retailers do with it.
Making Those Assets Work
The investments that strengthen a retailer’s position against agentic commerce are the same investments that strengthen its competitive position today. Clean customer data, structured product information, AI-powered personalisation, direct relationships that give customers reasons to come back. These deliver value now. They will compound in value as agent-mediated commerce grows.
The principle is straightforward: the deeper the data-led connection with your customer, the harder it is for an intermediary to displace you. Australian grocery retailers illustrate this well. Woolworths and Coles have spent years building sophisticated loyalty and data capabilities through competing with each other, operating AI-powered personalisation engines across millions of members. They also benefit from structural advantages: many-item habitual purchasing that is harder for agents to disintermediate, and very few major competitors. The combination of strategic investment and category dynamics makes grocery less vulnerable.
Outside grocery, the exposure is higher and the readiness more uneven. Myer’s commitment to its Myer One programme demonstrates the kind of strategic bet that the shift demands. The programme now has 4.7 million active members, with more than three quarters of all Myer sales linked to Myer One members. Members spend 2.8 times more than non-members. Under CEO Olivia Wirth, who brings a strong Loyalty background, Myer has positioned data-driven loyalty as a central business strategy, even during a challenging trading period. The programme won International Loyalty Program of the Year at the 2025 International Loyalty Awards, and the Myer One tag rate is now touching 80%. The broader business is in transition, but the loyalty metrics are heading in the right direction. The appointment of a loyalty specialist to lead the entire company signals where Myer sees its future.
For retailers who have not made equivalent investments, the opportunity is still there. Investing in first-party data collection through loyalty programmes and direct channels. Using that data to power personalised experiences on channels the retailer controls. Structuring product data so it is accurate, complete and machine-readable, which improves search performance, marketplace visibility, as well as agent discoverability. Building the kind of in-store and digital experiences that create genuine reasons for customers to return directly. None of this requires a grocery-scale technology budget. It requires a decision to treat the customer relationship as the strategic asset it is and to resource accordingly.
However, many retailers across fashion, specialty, homewares and general merchandise have a recognised brand and loyal following but have not built the underlying data and relationship infrastructure. An AI agent does not respond to brand sentiment. It evaluates data: product descriptions, pricing, availability, reviews, fulfilment reliability. A retailer with strong brand recognition but fragmented customer data and no structured product taxonomy may find that its brand carries far less weight when the customer’s agent is the one making the decision.
In most cases, brand alone is not a moat. (Unless you are in the market for a Hermès bag.) The moat is the combination of brand, data, direct relationships and the infrastructure to make those assets work in every channel, including the ones mediated by AI.
The Window Is Still Open. For Now.
The threat from agentic commerce to retail is small today but growing and definitely coming. GenAI shopping use grew 35% between February and November 2025, according to BCG. The infrastructure is being assembled now: Mastercard’s Agent Pay, Google’s Universal Commerce Protocol, endorsed by more than twenty major companies. The coordination layer is forming.
The customer relationship is not a legacy asset to be maintained. It is the most valuable thing a retailer owns in a world where AI agents owned by someone else are beginning to decide where consumers spend their money. The retailers who treat it as such, who invest in making those relationships deeper, more data-rich and more useful to both customers and the agents who will increasingly be acting on their behalf, are the ones who will retain influence as the coordination layer forms around them.
The retailers who wait will find themselves in the same position as hotels a decade into the OTA era: spending to reclaim what they could have protected.
Sources
E-commerce News Australia, “Leveraging loyalty: How are retailers approaching retention in 2024?”, June 2024
Marketplace Pulse, Amazon Third-Party Seller Share Data, Q4 2025
Pattern and Profitero, Amazon Brand Marketplace Survey, 2021
BCG, “First-party data benchmarks,” via Avaus, February 2025
BCG, “Consumers Trust AI to Buy Better. Brands Need to Move Quickly,” December 2025
Salesforce, “AI Agents Set To Boost Australian Shopper Experiences,” November 2024
Dsouza et al., “Decoding Gen Z: AI’s influence on brand trust and purchasing decisions,” Journal of Retailing and Consumer Services, March 2024
Loyalty & Reward Co, “Do Loyalty Programs Work? Just ask Myer,” October 2024
Gravitas, Myer International Loyalty Awards 2025 Results, May 2025
Channel News, “Myer Expands Loyalty Program,” October 2025
Myer Holdings, FY25 Results Release, September 2025