For more than two decades, keyword strategy shaped digital visibility. Businesses researched search intent, optimized landing pages, and built content around the exact phrases customers typed into a search box. That model assumed a human would read a query, scan a list of results, and click through to a website.  

Search behavior no longer follows that sequence for a growing share of queries. AI-generated answers, shopping assistants, retail media platforms, and recommendation engines increasingly retrieve structured product information instead of simply matching webpages to keywords. As these systems become part of everyday product discovery, structured product data is becoming as important to visibility as keyword strategy once was.

How AI Search Is Changing Product Discovery

Bain & Company's December 2024 survey with Dynata found that queries triggering an AI Overview end without a click 83% of the time. The same research found that 80% of consumers rely on AI-generated results for at least 40% of their searches, and organic traffic across many sectors has already declined 15–25%.

Adobe's transaction data, drawn from more than one trillion visits to U.S. retail sites, shows how fast this shift accelerated. Traffic from generative AI sources grew 1,200% between July 2024 and February 2025. By Q1 2026, that growth reached 393% year over year, following a 693% surge during the 2025 holiday season. The pace of growth is moderating only because the channel is now large enough that percentage gains are harder to post; the underlying volume keeps climbing.

This changes what "ranking" means. A page optimized around a keyword phrase can still rank in the traditional sense while never appearing in the answer a customer actually reads. An AI system extracts facts - price, availability, material, compatibility, dimensions - and assembles them into a response. Rather than relying primarily on keyword matching, AI systems evaluate structured product attributes alongside textual relevance before generating recommendations. A product's visibility now depends on whether that information exists in a format the system can extract with confidence.

Why Structured Product Data Matters Across AI Search, Shopping, and Retail Media

The same shift is reshaping the channels that have always driven ecommerce revenue, not just AI chat interfaces. Google Shopping, Performance Max, Microsoft Shopping campaigns, Amazon advertising, and retail media networks all rank and recommend products using structured attributes such as GTINs, categories, pricing, availability, and shipping data, alongside traditional keyword signals. Google has reinforced this direction through Merchant Center improvements and AI-powered Shopping experiences that read product feeds directly.

“Your product data is becoming your sales pitch” – Marcel Hollerbach, Co-founder and CIO at ProductsUp in Season 1, Episode 2 of Growth, Interrupted.

Automated bidding systems depend on the same inputs. Performance Max and comparable machine-learning-driven campaign types make targeting and placement decisions based on the product data available to them. An algorithm cannot compensate for an incomplete title, an inconsistent attribute, or an inaccurate availability status - it can only optimize against the data it has. Improving the underlying product feed therefore lifts performance across paid search, Shopping ads, retail media, and AI discovery simultaneously, rather than requiring separate optimization work on each platform.

Why AI-Referred Shoppers Are Becoming More Valuable

The conversion story is the more consequential part of Adobe's data. In March 2025, AI-referred traffic converted 38% worse than traditional channels such as paid search and email. By March 2026, that same traffic converted 42% better, with revenue per visit running 37% higher than non-AI sources. That reversal happened in twelve months.

It shows AI systems are not a peripheral discovery channel. They are becoming a primary purchase channel, and the shoppers arriving through them are further along in their decision when they land. A brand whose product data is incomplete or inconsistent is not losing a marginal traffic source. It is becoming increasingly difficult to compete in a channel that Adobe's data shows now delivers stronger engagement and higher revenue per visit than many traditional acquisition sources.

How AI Shopping Agents Are Reshaping Digital Commerce

The shift extends beyond answering questions to completing transactions. McKinsey estimates that agentic commerce - AI agents executing the full purchase process on a consumer's behalf - could redirect $3 to $5 trillion in global retail spend by 2030. Gartner projects that AI agents will intermediate more than $15 trillion in B2B purchases by 2028, and forecasts that 40% of enterprise applications will embed AI agents by the end of 2026, up from under 5% in 2025.

An AI agent completing a purchase cannot read a homepage, interpret brand photography, or infer meaning from a well-crafted headline the way a person can. It reads structured fields: name, brand, GTIN, price, availability, material, dimensions, and compatibility attributes. When an agent evaluates two competing products and one has a complete, machine-readable record while the other has gaps, the agent recommends the product it can verify. Incomplete data reduces an agent's ability to evaluate and confidently recommend a product. As AI purchasing systems become more sophisticated, comprehensive and verifiable product information is likely to become an increasingly important requirement for inclusion.

Product Data Strategy in Practice: From Keywords to Structured Data

Keyword strategy meant building content around the terms customers typed into a search box. Product data strategy means building a structured, verifiable record around every attribute an AI system needs to answer a customer's question with confidence.

Consider two versions of the same running shoe listing. The keyword-optimized version includes a description written around the phrase "best running shoes for trail use," supported by blog content targeting that term. The structured version includes the same core information, organized as explicit, machine-readable fields: weight in grams, drop height in millimeters, water-resistance rating, size range, colorway, and price with real-time availability, marked up in schema so a crawler or an AI agent can extract each fact directly. When a shopper asks an AI system "which trail running shoe under $150 handles wet terrain," the structured listing has an answer the system can lift and cite. The keyword-only listing has a phrase the system has to interpret.

This gap compounds at catalog scale. A brand with thousands of SKUs and inconsistent attribute coverage is not partially visible to AI systems. It is selectively invisible, product by product, wherever a field is missing or a value is ambiguous. In both cases, the product is identical. The difference lies in how easily an AI system can understand and verify the information.

Three Leadership Priorities for AI Commerce

Three shifts follow directly from this data.

1. Treat Product Data as a Strategic Growth Asset

Product information management, feed accuracy, and schema implementation now determine how a brand appears to search engines, Shopping platforms, and AI agents alike. These have traditionally sat with IT or merchandising teams, disconnected from search and content strategy. That separation no longer holds, given that structured data now drives the same visibility outcomes that keyword content once did.

2. Measure AI Search and AI Commerce Separately

Adobe's data shows AI-referred traffic behaving differently from every other channel - converting better, spending more time on-site, and generating higher revenue per visit. Attribution models built around organic and paid search alone will undercount where revenue is actually originating.

3. Invest in Product Data Quality Before Scaling AI

With Gartner and McKinsey both projecting trillions of dollars in commerce shifting to agent-mediated transactions within the next two to four years, and Adobe already documenting a sustained, multi-quarter surge in AI-referred traffic and conversion, the return on structured data investment is measurable today, not a forward-looking bet.

Conclusion: Product Data Is Becoming the Foundation of AI Commerce

Keyword strategy remains an essential part of digital growth, but it is no longer the only language that discovery systems understand. AI search engines, shopping platforms, and autonomous buying agents increasingly rely on structured product information to evaluate relevance, compare products, and generate recommendations.

As commerce becomes more machine-mediated, product data is evolving from an operational requirement into a strategic growth asset. Organizations that invest in complete, accurate, and machine-readable product information today will be better positioned to compete across search, retail media, and the next generation of AI-powered commerce.

The opportunity extends beyond improving visibility in a single channel. A well-structured product data strategy strengthens performance across AI search, Shopping platforms, automated advertising, retail media, and emerging agentic commerce, allowing the same product information to support discovery wherever customers choose to search, compare, and buy.

Key Takeaways

· AI-powered search, shopping assistants, and recommendation engines increasingly rely on structured product data alongside traditional keyword signals to determine product visibility.  

· Product feeds containing complete and machine-readable information improve performance across Google Shopping, Performance Max, retail media, AI search, and emerging agentic commerce.  

· AI-referred shoppers are becoming a higher-value audience. Adobe's data shows they now generate higher revenue per visit than traditional traffic sources, increasing the importance of accurate product information.  

· As AI agents begin comparing and purchasing products on behalf of consumers, complete product attributes become essential for evaluation and recommendation.  

· Organizations that treat product data as a strategic business asset, rather than an operational task, will be better positioned for long-term visibility across AI-powered commerce.

---

Curious how AI decides which products to recommend? Reach out to us.

Relevant Articles:

· Article: The Zero-Click Customer Journey: What It Means for Performance

· Article: How to Optimise for Generative AI Search: GEO vs. SEO

· Report: The State of Search 2026 and Beyond: How AI, Automation, and Commerce Are Reshaping Discovery

About Crealytics

Crealytics is an award-winning full-funnel digital marketing agency fueling the profitable growth of over 100 well-known B2C and B2B businesses, including ASOS, The Hut Group, Staples and Urban Outfitters. A global company with an inclusive team of 100+ international employees, we operate from our hubs in Berlin, New York, Chicago, London, and Mumbai.

‍