AI Agents Are Replacing Google Shopping: What Brands Must Know

Here is a number that should concern every e-commerce marketer: AI-assisted purchases grew 340% in 2025. Here is the number that should concern you even more: Google Shopping click-through rates declined 18% over the same period.

These two trends are not coincidental. They are cause and effect. As consumers increasingly delegate product research and purchasing to AI agents, the Google Shopping model — built on pay-per-click advertising where brands bid for visibility on a grid of product listings — is facing its most serious disruption since the platform launched.

This is not a speculative future scenario. It is happening now, and brands that fail to adapt will watch their customer acquisition costs rise while their market share erodes to competitors who understand the new landscape.

The Fundamental Shift: From Search-and-Browse to Ask-and-Buy

For two decades, the dominant e-commerce shopping model has been search-and-browse. A consumer types "wireless headphones" into Google, sees a grid of Shopping ads and organic results, clicks through to several product pages, compares options, and eventually makes a purchase. At every step, there are opportunities for brands to influence the decision — through ad placement, product imagery, pricing, reviews, and on-site experience.

AI shopping agents compress this entire journey into a single interaction.

The old model:

1. Consumer searches Google for product category
2. Consumer sees Google Shopping ads (brands pay per click)
3. Consumer clicks 3-5 product listings
4. Consumer visits product pages, reads reviews
5. Consumer compares options across tabs
6. Consumer adds to cart and checks out
7. Total time: 15-90 minutes

The new model:

1. Consumer tells AI agent what they need
2. Agent researches 50-100 products across dozens of retailers
3. Agent evaluates based on specs, reviews, pricing, availability
4. Agent presents 1-3 recommendations with rationale
5. Consumer approves (or agent auto-purchases if pre-authorized)
6. Total time: 30 seconds to 5 minutes

The implications for Google Shopping are severe. In the new model, nobody clicks on a Google Shopping ad. Nobody sees the sponsored product grid. The entire pay-per-click advertising infrastructure becomes irrelevant for agent-mediated purchases.

Why Google Shopping's Model Is Vulnerable

Google Shopping's business model is built on three assumptions that AI agents are systematically undermining.

Assumption 1: Consumers Start Product Searches on Google

Google's dominance in product search has been its moat. For years, "I need to buy something" meant "I'll Google it." But consumer behavior is shifting. According to recent data:

• 37% of consumers aged 18-34 now start product searches with an AI assistant rather than a search engine
• 24% of all consumers have used an AI agent to help with a purchase decision in the past 6 months
• Among consumers who have tried AI shopping, 68% report using it for their next purchase

This shift is accelerating. As AI agents become more capable and trustworthy — with features like autonomous purchasing and payment processing — the percentage of product searches starting on Google will continue to decline.

Assumption 2: Visibility Is Purchased Through Ad Bids

In Google Shopping, visibility is fundamentally an auction. The more you bid, the more visible your products. This pay-for-placement model generates billions in revenue for Google, but it has no equivalent in the AI agent ecosystem.

When an AI agent evaluates products, it does not consider which brands have paid for placement. It evaluates based on:

• Product-query match: How well does this product match what the consumer asked for?
• Data quality: How complete and structured is the product information?
• Price competitiveness: How does the price compare to alternatives?
• Review quality: What do authentic reviewers say about this product?
• Brand authority: How frequently is this brand mentioned and recommended across the web?
• Availability and fulfillment: Is the product in stock? How fast can it ship?

There is no "bid higher" lever in this model. Visibility is earned through product quality, data quality, and brand authority. This fundamentally changes the competitive dynamics of e-commerce advertising.

Assumption 3: Click-Through Is the Key Metric

Google Shopping optimizes for clicks. Brands pay per click. Google's algorithm maximizes total click value. But AI agents do not click. They query APIs, parse structured data, and make decisions programmatically. The concept of a "click" does not exist in agent-mediated commerce.

This means the entire measurement and optimization infrastructure built around Google Shopping — CTR, CPC, ROAS from Shopping campaigns — becomes less relevant as agent-mediated commerce grows.

How Each AI Platform Handles Shopping

Not all AI shopping experiences are the same. Understanding how each platform approaches commerce is essential for optimization.

ChatGPT Shopping (OpenAI)

How it works: ChatGPT Shopping is integrated directly into the ChatGPT conversation interface. When users ask about products, ChatGPT displays product cards with images, pricing, ratings, and buy links. For users with Operator access, ChatGPT can complete the purchase autonomously.

Where product data comes from: ChatGPT Shopping pulls product data from multiple sources — web crawling, product feed partnerships, Bing Shopping integration, and direct retailer API integrations.

Key visibility factors:

• Product data accessibility (schema markup, structured data)
• Review quality and quantity across platforms
• Brand mention frequency across authoritative sources
• Price competitiveness
• Content quality on product pages

Market share: ChatGPT Shopping is currently the largest AI shopping platform, with an estimated 45% share of AI-assisted product recommendations among US consumers.

Perplexity Buy

How it works: Perplexity integrates shopping into its search experience. When users ask product-related questions, Perplexity provides sourced recommendations with the ability to purchase directly from search results.

Where product data comes from: Perplexity primarily crawls the web for product information, pulling from retailer websites, review sites, comparison platforms, and product databases.

Key visibility factors:

• Content that directly answers purchase-intent queries
• Strong presence on review and comparison sites
• Detailed product pages with structured specifications
• Expert recommendations and editorial coverage

Market share: Estimated 15% of AI-assisted product recommendations, growing rapidly.

Google Gemini Shopping

How it works: Gemini leverages Google's existing Shopping Graph — a massive product database built from merchant feeds, web crawling, and structured data. Gemini can provide detailed product comparisons, price tracking, and increasingly direct purchase capabilities.

Where product data comes from: Primarily the Google Shopping Graph (fed by Google Merchant Center submissions), supplemented by web crawling and structured data extraction.

Key visibility factors:

• Google Merchant Center feed quality and completeness
• Google Shopping Graph inclusion
• Existing Google Shopping ad performance (currently still weighted)
• Product reviews aggregated in Google
• Schema markup on product pages

Market share: Estimated 20% of AI-assisted product recommendations, with significant potential for growth given Google's product data infrastructure.

Amazon Rufus

How it works: Rufus operates within the Amazon ecosystem, helping shoppers find products, compare options, and make purchase decisions within the Amazon marketplace.

Where product data comes from: Amazon's proprietary product catalog, customer reviews, Q&A sections, and product listing data.

Key visibility factors:

• Amazon product listing quality
• Amazon review quantity and quality
• A+ Content and Enhanced Brand Content
• Amazon advertising (Rufus incorporates some sponsored signals)
• Sales velocity and conversion rate

Market share: Estimated 18% of AI-assisted product recommendations (higher in categories where Amazon dominates).

Other Emerging Platforms

• Apple Intelligence Shopping: Expected to leverage Apple Pay and deep device integration for commerce. Likely to prioritize privacy-respecting product data.
• Meta AI Shopping: Commerce features across Instagram and WhatsApp, leveraging social commerce data.
• Specialized agents: Category-specific shopping agents for verticals like electronics, fashion, home, and automotive.

What Signals Agents Use Instead of Ad Bids

Since you cannot bid for placement with AI agents, what should you invest in instead? Here are the signals that drive agent recommendations, ranked by impact.

1. Product Data Quality (Highest Impact)

Agents need structured, comprehensive product data to make accurate recommendations. The brands with the best data get recommended most frequently.

What agents need:

• Complete technical specifications in structured format
• Accurate, real-time pricing and inventory
• Detailed compatibility and use-case information
• Clear product differentiation attributes
• Comparison-friendly data formats

How to improve:

• Implement comprehensive JSON-LD Product schema on every product page
• Use Shopify metafields or equivalent structured data on your platform
• Ensure your product feed (Google Merchant Center, etc.) is complete and up-to-date
• Add machine-readable specifications alongside human-readable descriptions

2. Review Sentiment and Quality (High Impact)

Agents perform sophisticated sentiment analysis on reviews. They do not just look at star ratings — they analyze the content of reviews to understand product strengths and weaknesses.

What agents evaluate:

• Overall sentiment distribution (not just average rating)
• Specific mentions of quality, durability, value
• Comparison statements ("better than X" or "worse than Y")
• Recency of reviews
• Review authenticity signals
• Response patterns from the brand

How to improve:

• Encourage detailed, descriptive reviews (not just star ratings)
• Respond to negative reviews constructively (agents notice this)
• Distribute reviews across platforms (Google, Amazon, Trustpilot, specialized review sites)
• Focus on review quality over quantity — 50 detailed reviews outweigh 500 "Great product!" reviews

3. Brand Authority and Mention Frequency (High Impact)

Agents build a model of brand authority based on how frequently a brand is mentioned, cited, and recommended across the web.

What builds authority:

• Expert recommendations and editorial mentions
• Inclusion in "best of" and comparison articles on authoritative sites
• Industry award recognition
• Social proof from thought leaders
• Consistent brand information across all platforms

How to improve:

• Invest in digital PR and expert outreach
• Seek product reviews from authoritative publications
• Participate in industry comparison roundups
• Maintain consistent NAP (Name, Address, Phone) and brand information across the web
• Create authoritative content that demonstrates product expertise

4. Price Competitiveness (High Impact)

Agents compare prices across all available retailers. Price is not the only factor, but it is always a factor.

What agents evaluate:

• Current price relative to competitors for the same product
• Price-to-feature ratio compared to alternatives in the category
• Historical pricing patterns (is this a fair price or an inflated one?)
• Total cost including shipping, taxes, and fees
• Discount and promotion authenticity

How to improve:

• Monitor competitor pricing in real-time
• Ensure price transparency (no hidden fees at checkout)
• Communicate value beyond price (warranty, service, quality)
• Consider price-matching policies that agents can factor into recommendations

5. Structured Data Completeness (Medium Impact)

Beyond basic product schema, comprehensive structured data helps agents make better recommendations.

What to implement:

• Product schema with full specifications
• Offer schema with shipping and return details
• AggregateRating and individual Review schema
• BreadcrumbList for category context
• FAQ schema for common product questions
• HowTo schema for setup/installation content

6. API Accessibility (Growing Impact)

As more agent interactions happen via API rather than web crawling, having accessible product APIs becomes increasingly important.

What to provide:

• Storefront API access (Shopify) or equivalent
• Real-time inventory endpoints
• Programmatic checkout capabilities
• Webhook support for real-time updates

Google's Response: AI Overviews and Shopping Graph

Google is not standing still. The company is aggressively integrating AI into its shopping experience.

AI Overviews in Shopping

Google's AI Overviews now appear for many product-related searches, providing AI-generated summaries and recommendations above traditional Shopping ads. This is Google's attempt to keep product searches within its ecosystem while embracing the AI-first experience consumers increasingly expect.

For brands, AI Overviews in Shopping are a double-edged sword:

• Positive: If your product is recommended in an AI Overview, you get prominent visibility without paying per click
• Negative: AI Overviews reduce clicks to individual product listings, potentially decreasing your Google Shopping ad volume

Shopping Graph Integration with Gemini

Google's Shopping Graph — which catalogs billions of products with rich structured data — is being directly integrated with Gemini's AI capabilities. This gives Gemini Shopping a significant data advantage over competitors like ChatGPT Shopping and Perplexity Buy.

For brands, this means your Google Merchant Center feed is more important than ever. It feeds directly into both traditional Google Shopping ads and Gemini's AI shopping recommendations.

Performance Max and AI Bidding

Google has updated Performance Max campaigns to factor in AI-driven shopping interactions. Your Performance Max campaigns now optimize for visibility across traditional Shopping, AI Overviews, and Gemini interactions. However, the measurability of AI-driven conversions within Google Ads remains limited.

Budget Reallocation: Shifting Google Shopping Spend

The question every e-commerce marketer is asking: should I cut my Google Shopping budget? The answer is nuanced.

The Recommended Approach

Do not abandon Google Shopping. It still drives significant revenue and will continue to for years. But begin strategic reallocation:

Year 1 (2026): Diversify 10-20% of Google Shopping budget

Reallocate toward:

• Structured data optimization (5-8% of budget): Invest in comprehensive schema markup, metafield completion, and product data quality across your catalog
• Review strategy (3-5%): Invest in review generation, review platform diversification, and review response programs
• AI visibility monitoring (2-4%): Tools and processes to track your brand's visibility across AI platforms
• Content for AI consumption (2-5%): Expert content, comparison data, and authoritative product information that agents use to build brand authority

Year 2 (2027): Expand to 20-35% reallocation

As AI shopping adoption accelerates:

• Agent-specific optimization: Direct integrations with agent platforms, API development, agent-compatible checkout flows
• AI advertising: Sponsored placements within ChatGPT Shopping, Perplexity Buy, and other platforms that offer advertising
• Brand authority building: Expanded PR, expert partnerships, and content programs that build the authority signals agents value

Year 3 (2028): Full omnichannel AI-plus-traditional budget

By 2028, expect 30-50% of product discovery to be agent-mediated. Your budget should reflect this reality.

Products and Categories Where Agents Win Fastest

Not all product categories are equally affected by the shift to agentic commerce. Agents are gaining share fastest in:

• Electronics and technology: Specification-driven purchases where agents excel at comparison
• Household consumables: Routine replenishment purchases that agents can automate
• Commoditized products: Products where brand matters less than price and availability
• Research-heavy purchases: Products where consumers traditionally spend significant time comparing

Agents are gaining share more slowly in:

• Fashion and apparel: Where visual aesthetics and personal style still drive decisions
• Luxury goods: Where brand experience and emotional connection matter
• Gifts: Where personal knowledge and creativity influence choices
• Impulse purchases: Where agents cannot replicate the "I saw it and wanted it" dynamic

Measuring the Shift in Your Category

Track these metrics monthly to understand how quickly agents are disrupting your specific category:

The New Playbook: Winning in Both Worlds

The brands that will thrive through this transition are those that optimize for both traditional Google Shopping and AI agent visibility simultaneously. Fortunately, many optimizations serve both channels.

Optimizations That Serve Both Channels

• Comprehensive product data: Better data improves Google Shopping ad performance AND agent recommendations
• Strong reviews: Reviews boost Quality Score in Google Shopping AND agent trust
• Competitive pricing: Price competitiveness drives Google Shopping clicks AND agent preference
• Complete Merchant Center feeds: Feed quality improves Shopping ad eligibility AND Gemini Shopping visibility

Optimizations Specific to AI Agents

• Machine-readable specifications beyond what Google Merchant Center requires
• API accessibility for direct agent interaction
• Cross-platform brand consistency that builds authority signals
• Expert content and editorial coverage that agents reference for brand authority
• Structured comparison data that helps agents evaluate your products against competitors

Optimizations Specific to Google Shopping

• Bid optimization for maximum ROAS in declining-CTR environments
• Creative testing for product images and titles in Shopping ads
• Negative keyword management for Performance Max campaigns
• Audience targeting for high-intent shoppers who still browse Google Shopping

What to Do This Quarter

Here is a concrete action plan for the next 90 days:

1. Audit your AI visibility: Query ChatGPT, Claude, Perplexity, and Gemini for products in your category. Document where you appear and where you don't.

2. Complete your structured data: Ensure every product has comprehensive JSON-LD schema markup with Product, Offer, AggregateRating, Review, and shipping/return data.

3. Enhance your product feed: Go beyond Google Merchant Center minimum requirements. Add detailed specifications, compatibility data, and use-case descriptions.

4. Invest in reviews: Launch or enhance a review generation program. Focus on platforms that AI agents reference — Google Reviews, Trustpilot, and category-specific review sites.

5. Set up AI visibility monitoring: Establish a regular cadence for testing your products against AI agents. Track changes monthly.

6. Begin budget diversification: Allocate 10% of your Google Shopping budget toward AI visibility optimization activities.

7. Brief your team: Ensure your marketing team, product team, and leadership understand the agentic commerce shift and its implications for your business.

The Google Shopping model is not dying tomorrow. But it is being disrupted, and the disruption is accelerating. The brands that prepare now — investing in the data quality, brand authority, and technical infrastructure that agents value — will own the next era of e-commerce. Those that cling to the bid-for-clicks model will find their customer acquisition costs rising while their market share falls.

The agent revolution is not coming. It is here.

Want to understand your brand's AI shopping visibility and build a transition strategy? Contact AdsX for a free AI visibility audit. We help brands win in both traditional and AI-powered shopping channels.