Google vs ChatGPT. The E-Commerce Battle Is On

For now, Google is still testing this feature, so it’s only available in the United States and requires you to be logged into your Google account. If you’re outside the US, you can still access it by using a VPN.

To try AI-assisted shopping, simply type your query into the search bar and switch to the AI Mode tab.

Alternatively, you can go directly to the chat widget by clicking on the AI Mode button.

If AI Mode doesn’t appear for you, try enabling it manually by visiting: google.com/aimode

To try ChatGPT’s version, the process is even simpler. Just go to the chat and ask your shopping-related question. It’s available globally, with no regional restrictions.

We covered ChatGPT’s shopping assistant in detail in our previous article.

In this piece, we’ll focus more on Google’s version and what it means for merchants and platforms.

So how does Google’s AI decide what to recommend? Whenever a user requests a product recommendation, the system performs two main tasks:

• 🔸 It identifies the most relevant products to suggest.
• 🔸 It determines which stores are the best sources for purchasing those products.

This approach is quite similar to how ChatGPT’s Shopping Assistant works, which we explored in detail in the previous article. However, Google uses a different set of tools and methods. Instead of Microsoft Bing (used by ChatGPT), Google relies on its own search engine to gather information from across the web. It searches for content related to your query and draws from sources such as:

• ▪︎ Articles like “Best products for X” or “Top items in Y category”
• ▪︎ Individual product reviews
• ▪︎ YouTube videos (it can read video transcripts)
• ▪︎ Websites like Reddit and Quora
• ▪︎ Social platforms like Twitter and Facebook
• ▪︎ User-generated product reviews and ratings
• ▪︎ And more

These are the kinds of reputable sources you would typically see on the first or second page of search results.

Google then combines this information with data from its Shopping Graph, and in some cases, it even visits specific product detail pages (PDPs) on merchant websites to verify or extract relevant information. The result is a curated list of recommended products and places to buy them.

The Shopping Graph itself isn’t new. It’s been a core part of Google’s e-commerce infrastructure since 2012. Think of it as a massive product catalog that includes over 50 billion listings from both global retailers and local businesses. Each listing may include reviews, pricing, availability, colors, and other relevant attributes.

According to Google, the Shopping Graph pulls its data from a variety of sources, including:

• ▪︎ YouTube videos
• ▪︎ Manufacturer websites
• ▪︎ Online stores and product detail pages
• ▪︎ Google Merchant Center
• ▪︎ Google Manufacturer Center
• ▪︎ Product reviews and test results
• ▪︎ And many other sources

The AI model that powers this recommendation engine is based on a large language model (LLM), specifically Google Gemini, which handles the natural language processing and decision-making behind the assistant’s suggestions.

Now let’s run a field experiment when it comes to helping with real-world shopping questions. For this test, we used ChatGPT-4o with a Plus subscription, with memory turned on to make use of all available personalization and context features. This is the latest version as of May 22, 2025.

On the Google side, we used AI Mode on Gemini 2.0 through a personal account that’s seen plenty of regular use – giving it a good amount of context from past searches and activity.

The idea here is to run the exact same prompt through both tools and see which one delivers more useful results. We’re looking at how they handle the process in general, including product suggestions, price filters, delivery timelines, and what else we will find.

Before jumping into the results, it’s worth calling out the user experience.

ChatGPT, being rooted in a traditional chat format, still sticks to a pretty basic vertical layout. While that works for conversations, it’s not really optimized for shopping or search tasks. The interface hasn’t been updated (yet?), so on larger screens it feels a bit off – most of the space on the sides goes unused, and the chat itself is confined to a narrow column that feels more suited for reading than browsing. When OpenAI first teased the shopping assistant, we saw a preview with product cards (everyone reposted those Delongi coffee makers) on the right, but that version isn’t live just yet.

On the other side, Google’s AI mode makes the most of its layout. It looks and feels more like a modern productivity or search tool, with navigation on the top and left, a centered chat window that’s easy to read, and product recommendations neatly displayed on the right. It’s clear Google leaned into its design experience to make everything feel more interactive and efficient.

Here’s the first prompt we tested:

“I’m thinking about starting to run outdoors in my neighborhood now and then, especially when the weather’s nice. I’m looking for a pair of running sneakers that are reasonably priced, preferably under $160, and comfortable for wide feet. Can you help me find the best options that can be delivered to my place in NYC, 10009, within three days?”

Now, let’s get into the actual product recommendations.

ChatGPT came back with five sneaker options, each one labeled with a tag you’d normally see in a product roundup, like Best Overall, Budget Friendly, Maximum Cushioning, Lightweight Choice, and Responsive Ride.

It was clearly going for that “review article” vibe, but there were a few misses. For example, the Budget Friendly pick wasn’t actually the cheapest one, and some options were priced above the $160 limit we set. So while it tried to follow the brief, it didn’t fully stick to the criteria.

The product recommendations showed up as cards embedded in the chat, which you can scroll through horizontally. Each card included the same tag from the review summary, and at the bottom of the response, ChatGPT listed the sources it pulled the info from, which is a nice touch for checking credibility.

Google

Google’s AI Mode took a more structured approach. It returned two top product picks, clearly marked and explained, along with three additional suggestions labeled as Other options. For those three, it added a note saying that buyers should double-check the availability of wide sizes.

All five products appeared as cards right in the chat, and on the right side of the screen, there was a list of 12 source sites it pulled data from. The cards were well-designed, each one included a product image (auto-generated with white background), star ratings with the number of reviews, and current pricing, including any discounts pulled from what looked like Google Merchant feeds.

The text summaries for the top two picks were more detailed, covering pricing, comfort, and delivery. The Other options got shorter blurbs but still included the key info.

What really stood out was how Google also recommended three retail stores beyond the usual Amazon and Zappos links. These local stores were suggested as better bets for fast delivery. Each card included the store’s name, rating, reviews, hours, and address, and clicking them opened the exact location in Google Maps, complete with directions and user reviews.

At the end, the assistant added an Important Note section with tips on what to look for in wide shoes, reminders to check return policies, and a suggestion to get a professional foot measurement if possible.

It even wrapped up the answer with a local map view, showing nearby stores that carried the recommended models.

Back to ChatGPT

Looking at ChatGPT’s sources, you can clearly trace how it built its response. First, it pulled insights from review-style articles to figure out which sneakers to recommend. Then it used its built-in search mechanism (powered by Bing, as we covered in our previous article to find live product listings).

It checked merchant websites for pricing, availability, and shipping to the NYC zip code we gave, and also referenced manufacturer websites to get official product descriptions. Along the way, it pulled in some general content sources, like Reddit threads, to add broader context and user opinions.

Google

Google’s AI Mode followed a similar path but leaned more heavily on major retail platforms like Amazon and Zappos. Instead of pulling from discussion forums or community sites like Reddit, it focused on official product listings and merchant reviews, which it seems to access directly and quickly.

The product cards looked like (obviously) they were powered by existing integrations with Google Shopping data, so details like pricing, availability, ratings, and images were right there, no need to dig through links. It felt more like a native shopping experience than a generated response.

ChatGPT

When you click on a product card or the product name in either tool, you get a detailed view on the right-hand side of the screen. In ChatGPT, the expanded view includes a product image with an auto-generated background, color options, star ratings with unknown source (based on third-party data with a disclaimer that the info may change), and pricing with Buy buttons linking to merchant sites. It also shows a list of detailed reviews, with sources listed below, giving it the feel of a well-sourced product roundup, even if it’s not always clear where the ratings are pulled from.

Google

With Google AI Mode, the product detail experience is on a whole different level. It starts with a more intuitive interface, including familiar navigation like Share and Options buttons right at the top. You get a 360-degree auto-generated view, a wider selection of product images for each color variation, and clear, clickable ratings and reviews linked directly to their sources. No surprise here, Google already owns much of this data through its Shopping and Merchant integrations.

Pricing is displayed with options and ranges, and while a price tracking feature is announced, it wasn’t active during our test. Still, what really stands out is the richness of the content: real-life photos from buyers, video reviews, star ratings, and customer-written overviews all presented in one place.

Google doesn’t stop there. You also get suggestions for related products and prompts for other search paths, which brings you back into a more traditional (but supercharged) Google Shopping experience. For merchants, this opens up a lot of visibility and engagement opportunities.

Summary

Both ChatGPT-4o and Google AI Mode (Gemini) offer solid shopping assistance, but they serve different experiences. ChatGPT provides a review-style response with curated picks and source links, though it occasionally drifts from the user’s criteria and lacks interface depth. Google AI Mode, on the other hand, delivers a far richer, more interactive experience, leveraging its Shopping ecosystem to offer detailed product views, real customer content, local store options, and seamless navigation. When it comes to usability, data depth, and retail integration, Google clearly leads, while ChatGPT shows promise as a conversational shopping guide.

This is another valuable feature from Google that helps users shop with more confidence. ChatGPT and other assistants don’t currently offer anything similar, which makes this especially noteworthy.

Check out this article by Google to learn more about how Virtual Try-On works and why it matters for online shopping.

Google doesn’t plan to stop at simply recommending what to buy. The next step is full transaction support, allowing the assistant to purchase items on your behalf from the websites you choose. That means skipping the usual steps of visiting the site, adding items to your cart, and going through the checkout process. Google will handle all of that for you.

On top of that, it will be able to monitor prices and stock availability, then complete the purchase automatically once your criteria are met. Naturally, the expected payment method is Google Pay.

According to Google, this assisted purchase feature is expected to roll out in the coming months.

Google isn’t alone in this pursuit. ChatGPT, Perplexity, PayPal, Visa, and other major players are also moving toward enabling seamless online transactions through AI-powered agents.

From a technical and business perspective, this shift requires new protocols to support bot-driven transactions, potentially including those that involve user logins to access personalized offers, or discounts. Merchant websites will need to support these types of integrations to fully participate.

In short, this is the next evolution of what was once known as eProcurement, a concept historically used in B2B e-commerce, and now making its way into mainstream consumer shopping.

The real winner here is us, the customers. But on the other side of this transformation, Google is positioned to become the ultimate winner as a platform for AI-assisted product discovery and automated purchases.

Why? Simply because Google has a much larger active user base than ChatGPT, Perplexity, Bing, and other competitors. We’re talking about billions of daily users across the globe.

Google also has access to an enormous amount of user data, gathered from services like Gmail, Google Drive, Search, YouTube, Maps, and even Google Wallet. This includes everything from your search history to your travel routes, viewing habits, and past purchases. While not all of this data is used for AI training yet, that might change soon — just wait for the next update to Google’s Terms of Service 🙂

In addition, Google has been a major player in e-commerce for years through services like Google Shopping and Google Ads. It has developed what is arguably the most comprehensive product catalog in the world, known as the Shopping Graph.

And let’s not forget Android, the operating system that powers the majority of mobile devices worldwide.

Altogether, these assets give Google a unique competitive advantage in both reach and the quality of personalized, commerce-driven recommendations it can deliver to its users.

Optimizing for Google is nothing new for online stores. In fact, most have already been doing this in one form or another for years. The process won’t be entirely different from existing SEO and e-commerce best practices, and much of it closely aligns with the optimization strategies we previously shared for ChatGPT.

That said, here are key actions a business can take to increase the likelihood that its products are recommended by Google, ideally more often than those of competitors. These suggestions are based on what we currently know about Google’s recommendation engine, combined with our hands-on experience in the e-commerce space:

Product and Brand Exposure

Content matters as never before. It is vitally important to have your products and brand featured in both onsite content (your own website) and offsite content (such as online magazines, blogs, YouTube, Facebook, Reddit, etc.), especially in articles that present you as a good solution to a specific problem. Articles focused on topics like “best,” “top,” or “how to” are frequently used by ChatGPT when recommending products.

These types of posts are especially valuable because they directly answer common user questions and support the decision-making process. To be effective, the content should be full-length, well-written, and offer thorough coverage of the topic. Ideally, it should appear on reputable websites that rank well in search results for relevant keywords. Creating content for real people not only improves engagement, it also makes it more accessible and useful to large language models like Gemini.

Google SEO Optimization

As shown in the examples above, most of the sources Google pulls information from are reputable websites and online stores — the kind that typically appear on the first page of Google search results for relevant queries. This means ranking and reputation still play a significant role.

However, some studies suggest that there isn’t a direct or guaranteed link between ranking on the first page and always being recommended by the AI. The relationship exists, but it is not strictly linear.

Submit Your Product Feed to Google Shopping

If you haven’t already submitted your product feed to Google, this is a critical first step. It’s essential for gaining visibility in product recommendations, as the product feed is one of Google’s primary sources of information.

Equally important is the quality of the data you provide. Your feed should be accurate, complete, and well-structured, with each product clearly and properly described. High-quality data increases the likelihood that your products will be correctly indexed and favorably recommended.

Stay Updated on Google’s Auto-Buy API

Keep an eye on announcements from Google regarding the release of its auto-buy API. Once it becomes available, ask your development team to implement it on your website as soon as possible.

It’s important to note that the API will likely evolve over time. Google’s team will continue refining it based on real-world usage and feedback, so integration will not be a one-time task, especially in the early stages.

Ideally, major e-commerce platforms such as Adobe, Shopify, Shopware, BigCommerce, and others will offer built-in support for this protocol. This would make integration much easier, potentially requiring just a few clicks within your platform’s backend.

Implement Google Pay on Your Website Today

Google Pay is important not just because it will likely serve as the default payment method for Google’s upcoming auto-buy feature, but also because it is already one of the most preferred payment options among users. Its value is clear — both for current convenience and future readiness.

Optimize Your Product Pages (PDPs)

It’s easy to understand why it’s important to have well-structured, information-rich product detail pages. They serve both Google’s crawling bot and real users who land on your site.

Remember, you are not only optimizing for a large language model. Your human visitors are also unlikely to browse your site deeply, so each product page needs to do the heavy lifting.

Here are key elements to include:

• 🔸 High-quality images
• 🔸 Clear, accurate pricing
• 🔸 Well-written product descriptions
• 🔸 Explain when, where, and by whom the product is best used, based on common customer needs.
• 🔸 Transparent delivery terms and shipping costs, if possible
• 🔸 User-generated content such as reviews or ratings
• 🔸 Clear on-page structure using headers, meta tags, bullet points, and lists
• 🔸 Product Schema Markup
• 🔸 Details on availability, sizing, colors, and other options
• 🔸 A clearly stated return policy

Additionally, since it is likely that potential shoppers will not explore the rest of your website in depth, it is smart to do the following:

• 🔸 Provide clear information about your brand, physical store locations (if any), and other trust-building details.
• 🔸 Include upsell and cross-sell options directly on product detail pages (PDPs), as this becomes even more important./li>

Most of what is mentioned above falls under basic website hygiene and best practices, which ideally should already be in place.

User Generated Content

Customer reviews and ratings are one of the factors Google uses when recommending products. They help determine product relevance and highlight specific characteristics, such as whether an item fits true to size, is durable, well-made, or matches the expected color.

Ensure All Pages on Your Website Are Easily Accessible

If a page cannot be accessed by clicking through a clear series of links or buttons, there is a good chance Google’s bot will have trouble discovering it.

Effortless Checkout

The importance of having an easy, clear, and trustworthy checkout experience cannot be overstated, especially one that avoids surprising customers with unexpected charges, shipping conditions, or other last-minute changes.

Imagine a buyer sees one price or shipping offer in Google, but encounters something different at checkout. With an abundance of other offers, it becomes even easier for them to abandon the purchase and look for another seller.

It’s safe to assume that pay-per-click (PPC) ads will make their way into Google’s AI-powered shopping experience sooner rather than later. Advertising has always been a core part of Google’s revenue model, and this new format will likely follow that pattern.

As for other platforms like ChatGPT, it’s also likely that we’ll see some form of shopping ads introduced there in the future. Monetization will be a natural next step as these assistants evolve.

It’s essential to embrace what’s coming and begin preparing for it now. In this article, we’ve shared actionable steps merchants can take today to improve their visibility in AI-powered product recommendation engines.

When it comes to which platform might dominate, Google appears to have a significant edge. With its massive user base across devices, deep integration into daily life, and vast access to shopper data, it is destined to lead. And let’s be honest — Google has a strong track record of delivering top-tier products to market. Can we say the same about the competition?

While AI-assisted shopping will play a major role, direct traffic to your website will still matter. Offline sales channels will remain relevant as well. That’s why investing in your website and your overall online presence is just as important as ever, possibly even more so.

P.S. We have to admit, we were skeptical about Google’s approach at first, especially when we reviewed ChatGPT’s shopping capabilities. But we were wrong. Well done, Google!