Max Sinclair is the CEO and Co-Founder of leading Agentic Commerce Optimisation and AI Visibility Startup Azoma.ai.
E-commerce brands have long optimized for search rankings, marketplace placement and paid media.
Now, another layer is influencing what consumers see and buy: AI agents.
When a shopper asks Alexa for Shopping, or when they use Walmart Sparky, ChatGPT or Gemini for the best product for a particular need, the system is doing more than retrieving a list. It is interpreting the question, assessing products against the request, weighing evidence from different sources and deciding which options it can confidently recommend.
That creates a new optimization problem for brands. Agentic commerce optimization (ACO) is the process of improving how products are discovered, understood, evaluated and recommended by AI shopping agents.
As the CEO and co-founder of Azoma, where we help leading brands with ACO and generative engine optimization (GEO), we analyze millions of AI shopping interactions to understand what influences product recommendations.
While ACO builds on search engine optimization (SEO) and answer engine optimization (AEO) principles, I’ve found many brands still struggle with applying the lessons from those disciplines to systems that can interpret intent and make recommendations.
From Search Rankings To Recommendations
Traditional e-commerce gives brands a clear objective: improve rankings, increase visibility and drive shoppers to a product page.
AI agents compress that journey. A shopper might ask, “What is the best moisturizer for sensitive skin?” The agent has to interpret the requirements, understand the products available, determine which products fit and construct a recommendation.
A brand can have a strong product, a well-optimized website and good marketplace placement while still being absent from the answer. The agent may rely on information beyond the product catalogue, including product specifications, reviews, editorial coverage, affiliate roundups, retailer information and other third-party sources.
ACO brings together the work required to make a product understandable, relevant, supported by credible sources and accurately represented.
What does agentic commerce optimization involve?
Based on where I’ve seen brands struggling, I believe there are five areas brands need to consider to develop their ACO strategy: completeness, context, citations, correctness and customer acquisition.
1. Completeness: Can the agent understand your product?
AI systems need enough information to understand a product and when it should be recommended.
A catalogue can contain thousands of SKUs, yet important information may be missing, inconsistent or difficult to compare. Ingredients, dimensions, use cases and pack sizes all affect whether a product matches a shopper’s request.
For AI shopping, this matters when shoppers combine several requirements. “Which running shoes are best for long-distance training, under $150, with a wide fit?” requires more than just a product name.
Brands must audit their digital shelf and ensure the attributes AI systems are likely to use are complete and consistent.
2. Context: What questions is your brand ready to answer?
Consumers rarely describe their needs using the same language brands use in marketing. A brand may describe a product as “advanced hydration technology,” while a shopper asks, “What should I use if my skin gets dry?”
Brands need to understand the questions consumers ask and create useful, brand-compliant content that directly answers them. Product pages remain important, but useful content connects products to real-world needs.
3. Citations: Where does the agent get its confidence?
An AI system can use information from a brand’s own website, but it also looks elsewhere before making a recommendation. Third-party sources provide additional evidence about products and brands.
My company recently performed an observational analysis of millions of prompts and responses across all brands and product categories on the Azoma platform during Q2 2026. While it should be noted that this is not a random sample, the analysis found that 73% of the citations observed in Alexa for Shopping responses came from affiliate sites, 16% from earned media and 11% from brand-owned website content.
This makes off-page visibility part of ACO. Earned media, affiliate coverage, reviews, forums and other relevant sources all provide evidence about products and brands.
4. Correctness: Is AI getting your product right?
AI systems get things wrong. A product can have accurate information in its catalogue while an AI assistant describes it incorrectly. It may confuse variants, attribute an ingredient to the wrong product or repeat outdated information.
Brands, therefore, need to query AI systems systematically rather than assuming correct source data results in correct recommendations.
Does the model understand the product? Does it associate the right attributes with the right SKU? Does it recommend the product for situations where it belongs?
5. Customer acquisition: Does visibility become revenue?
The first four areas converge on whether visibility in an AI answer creates commercial value.
Traditional e-commerce measurement is built around impressions, rankings, clicks and conversions. AI shopping adds another stage. A consumer may never see a conventional search result or visit a category page. An agent may recommend a small number of products and take the consumer directly toward a purchase.
Brands need to understand how often they are surfaced, which questions trigger recommendations and how recommendations translate into traffic and sales.
Why ACO Extends Beyond The Shopping Cart
The same principles apply beyond shopping assistants. Alexa for Shopping and Walmart Sparky are explicitly designed around commerce, while ChatGPT and Gemini can influence product discovery across a broader range of questions.
These are shopping questions even when they are not phrased as searches for a particular product. The systems answering them are making judgements about relevance, evidence and trust.
That is why ACO and GEO are increasingly connected. GEO addresses how a brand appears across AI-generated discovery more broadly, while ACO applies those principles specifically to the product and purchasing journey.
The New Product Discovery Layer
This does not mean abandoning SEO, marketplaces or traditional product optimization. It means adding another layer to discovery.
But brands also need to understand what happens when an AI agent sits between the consumer and those assets.
The agent decides which information to retrieve, which sources to trust and which products fit the question. The wider digital presence therefore becomes part of product discovery.
ACO is about preparing for that decision: completing the data, building the context, strengthening the sources, monitoring correctness and connecting visibility to acquisition.
AI shopping will evolve, but the requirement is clear: Brands must be understandable enough to be considered, relevant enough to be recommended and credible enough to be trusted.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
