ChatGPT is using Google shopping: what it means for performance marketing

Artificial intelligence is beginning to influence how consumers discover products online. Many people now ask AI tools questions such as “What are the best running shoes?” or “Which laptop should I buy?” before they even visit a search engines or retail sites.

For marketers, this raises an important question. If AI tools are shaping early product discovery, where are those recommendations coming from?

Early research suggests these AI responses may rely on existing ecommerce data sources. A recent Semrush analysis of ChatGPT recommendations found evidence that ChatGPT performs encoded queries to Google Shopping while generating product recommendation lists. The study observed that the products suggested by ChatGPT frequently matched items ranked highly in Google Shopping results. Around 75% of recommendations corresponded with products appearing within the top three positions.

This does not mean that AI systems always depend on Google Shopping. However, the research highlights an important possibility; some AI-generated product recommendations may draw from structured commerce data that already powers traditional search and retail platforms.

Why structured product data may influence AI answers

Large language models are designed to generate and interpret text. They are less suited to maintaining accurate information about stock levels, pricing or retailer availability.

Commerce platforms solve this problem with structured product data. Product feeds and shopping databases contain standardised attributes such as price, availability, retailer and reviews. This type of structured information can help AI systems provide recommendations that are practical and current.

In practice, conversational AI may combine two sources of information. It uses the open web to understand a topic and structured commerce data to identify relevant products.

For marketers, this suggests that product visibility across shopping ecosystems could influence how brands appear in AI-generated responses.

AI discovery and the changing purchase journey

AI tools are increasingly part of the research phase of online shopping. Instead of scanning multiple search results pages, users may begin with a shortlist suggested by an AI assistant.

This shift means some of the early evaluation work is done before users reach traditional search results or ecommerce category pages. In practice, AI tools summarise information from multiple sources and present a small number of options for further explanation.

Early data suggests this traffic can be highly engaged. Analysis of retail behaviour using Adobe Analytics found that visitors arriving from AI search tools often spend longer on site and view more pages than typical search visitors, according to reporting on Adobe Analytics data showing AI search traffic engagement.

This suggests that AI assistants may complement existing search behaviour rather than replace it. Users may begin with an AI-generated shortlist and then continue their research across search engines, review sites and ecommerce platforms.

For ecommerce brands, the implication is that discovery can now occur across several connected environments. These include search engines, marketplaces, retail media platforms and increasingly AI interfaces.

Implications for paid media and retail media strategy

These developments are taking place alongside broader changes in digital advertising. UK digital advertising expenditure is expected to approach £45 billion by 2026, with strong growth coming from retail media and data-driven advertising formats, according to the IAB UK digital ad spend forecast for 2026.

Retail media networks, product feeds and marketplace ecosystems already play a central role in ecommerce marketing strategies. If AI assistants reference structured commerce data, these environments could indirectly influence which products appear in AI-generated recommendations.

Analysts also expect investment in advertising linked to AI-driven search experiences to increase as platforms experiment with integrating ads into conversational interfaces. This trend is discussed in Reuters' coverage of forecasts for AI-driven search advertising growth.

At the same time, competition in lower-funnel performance channels remains strong as brands continue to prioritise measurable acquisition activity. Recent benchmarks highlighting this pressure appear in the Skai digital marketing quarterly trends report.

Taken together, these trends suggest that AI discovery will interact with existing marketing infrastructure rather than replace it.

What marketers should take from this

The current evidence does not suggest that AI replaces traditional search or ecommerce channels. Instead, it indicates that AI discovery may rely on many of the same data sources that underpin digital commerce today.

Product feeds, retail media visibility and reliable product information may influence how brands surface within AI-generated product recommendations. At the same time, familiar drivers of conversion remain important. These include brand reputation, price perception, customer reviews and the quality of the user experience.

AI assistants may change how consumers begin their research. However, the broader ecosystem of search, ecommerce platforms and performance marketing continues to shape how products are discovered and purchased.

For marketers, understanding how these systems interact will become increasingly important as AI discovery tools develop.

Navigating AI-driven discovery

At Planit, we help brands understand how changes in search, retail media and emerging AI discovery tools affect performance marketing strategy.

If you want to explore how evolving discovery channels may influence your acquisition strategy, our team would be happy to share what we are seeing across paid media, affiliates and retail ecosystems.

Sources

• Semrush – ChatGPT Searches Google Shopping to Create Its Recommendations

• IAB UK – UK Digital Ad Spend Forecast

• Skai – Digital Marketing Quarterly Trends Report

• The Verge / Adobe Analytics – AI search traffic engagement data

• Reuters – AI search advertising forecast