AI shopping beat search on Prime Day, quietly threatening Amazon's discovery business model

AI shopping beat search on Prime Day, rattling Amazon

AI shopping referrals converted better than any other channel on Prime Day for the first time, and that quietly threatens the discovery business Amazon and Google built. Amazon built Prime Day to keep shoppers inside its own marketplace. This year, AI shopping sent its best customers in from outside: a chatbot made the recommendation. US shoppers spent a record $26.4bn across retail sites during the four-day Prime Day event, according to Adobe Analytics. The more interesting number is not the total. It is where the most valuable buyers came from. For the first time, people who arrived at a retailer from an AI assistant were the most likely to actually buy, GeekWire reported. Shoppers referred by AI chatbots were about 40 per cent more likely to complete a purchase than those who came from search, email, or social media, Forbes noted. A year ago, the same traffic was the worst-converting channel of the lot. That reversal, inside 12 months, is the real story. The numbers behind the shift Adobe tracks billions of visits to US retail sites, and its Prime Day readout was lopsided. On day one, generative-AI traffic to retailers nearly doubled year on year, up 98.3 per cent. The visitors did not just show up. They behaved like buyers rather than browsers. AI-referred shoppers spent 49.9 per cent longer on site and viewed 20.5 per cent more pages. They added items to their baskets at a 33 per cent higher rate than people from traditional sources. In other words, the channel that used to waste a retailer's time is now its sharpest. The pattern is bigger than one sale. Adobe data shows AI-referred traffic to US retail sites rose 393 per cent year on year in the first quarter. By March the channel was converting around 42 per cent better than non-AI traffic. Rewind to early 2025 and AI visitors converted roughly 38 per cent worse. The line has crossed. Why a referral source rattles Amazon Amazon's whole model rests on being the place where the product search begins. That is where its advertising business lives, and ads are now one of its fattest margins. The Adobe figures point to a different starting point. If shoppers increasingly begin with ChatGPT, Gemini, or Perplexity and only touch a retailer to check out, the discovery layer moves up the stack to the AI firms. So does the money attached to it. The assistant, not the marketplace, becomes the shop window. Amazon is not standing still. It is pushing its own Alexa+ assistant and its Rufus shopping bot, and it is rethinking how much it pays for the AI models behind them. The catch is that the Adobe data shows general-purpose chatbots already doing the top-of-funnel work, often before Amazon enters the picture at all. Search is no longer the front door For two decades the journey began on Google. The same pressure now bearing down on AI shopping is the one cracking Google's grip on search: people are asking an assistant instead of typing a query. Marketers have a name for the response. They call it GEO, and the pitch is blunt: AI search is the new SEO. For retailers the lesson is practical. Catalogues need to be legible to machines, because a chatbot can only recommend what it can read. Adobe notes that most retail sites are still not very machine-readable, which leaves a gap between the traffic that is coming and the stores ready to receive it. There is a wider commerce land grab here too. OpenAI has already told advertisers it is in the ad business now. A chatbot that can both recommend a product and take the payment starts to look a lot like a marketplace. The European angle One caveat for readers outside the US: the Adobe figures track American retail. Europe is not there yet. Adoption of shopping assistants is slower here, and the EU's Digital Markets Act is already reshaping how Google and Amazon can rank and surface products. If assistants do become the front door in Europe too, regulators who spent years prising open search will face the same fight one layer up. The firms that own the assistant would sit where the gatekeepers used to. That is the deeper point for European retailers and brands. Visibility is shifting from a search box they have learned to game to a model they cannot see inside. The AI labs are writing the rules of that new shelf now, mostly in the US, and mostly on their own terms. The caveats Two cautions are worth keeping in view. The data comes from a single vendor, Adobe, which sells both the analytics and the AI tools that benefit from this story. Independent numbers would help. The second caution is scale. AI referrals are still a small slice of total retail traffic, even after tripling. A channel can convert best and still be small, and a four-day sale is not a full year. The direction looks clear. The size does not yet. Still, the front door of ecommerce is quietly being rebuilt. For years the question was how to rank on Google or how to win the Amazon buy box. The new question is simpler and harder: when the assistant makes the recommendation, who owns the customer?

Prime Day shows how AI is changing shopping, testing Amazon's bet against ChatGPT and others

U.S. shoppers spent a record $26.4 billion across all retail sites during Amazon's four-day Prime Day event, and for the first time, the people most likely to complete a purchase were those who arrived from AI chatbots. It's the latest twist in a high-stakes bet by Amazon. The AI assistants now sending retailers their best-converting customers are the same ones Amazon has worked to keep away from its own store, hoping to keep shoppers coming directly to Amazon.com and using its own on-site AI assistant instead. Adobe reported over that weekend that visitors who clicked through to shopping sites from AI assistants were 40% more likely to make a purchase during the four-day event than those showing up through search, email or social media. AI still accounts for a small fraction of total shopping traffic, but a trend is starting to emerge. In the past, shoppers sent by AI were the least likely to buy, according to Adobe's data. The change suggests that ChatGPT, Claude, Gemini and others are becoming more effective at giving shoppers the information they need to buy with confidence. Those figures span all of U.S. retail -- "Prime Day" has become much more than a day, and much bigger than Amazon alone. The distinction matters, because Amazon has taken a different path than many of its rivals. While Walmart, Target and others have opened their catalogs to outside AI assistants, Amazon has kept them out. Agentic AI drives less than 1% of traffic across every major online store, but Amazon's share is the lowest of the group, at about 0.4%, according to J.P. Morgan data. That's by design: Amazon sued Perplexity, for example, over its browser that shopped on customers' behalf, and won a preliminary injunction barring the tool from the logged-in parts of its site, arguing that unauthorized shopping agents degrade a trusted experience. Perplexity is appealing. Amazon has separately blocked ChatGPT's crawlers from reading its listings -- even as it has begun buying ads inside ChatGPT to bring shoppers back, a move first spotted by Marketplace Pulse founder Juozas Kaziukėnas and reported by Business Insider and Modern Retail. On Amazon's most recent earnings call, in April, CEO Andy Jassy said the company was in talks with the AI companies to come up with a better experience between Amazon and third-party agents to "find something that works for customers and all the companies." In the meantime, Amazon is focusing on its own AI assistant. The tool -- launched as Rufus and folded in May into a service called Alexa for Shopping -- has drawn more than 250 million users, with monthly users up more than 115% over the past year, the company said. Customers who use it while shopping are more than 60% more likely to buy, and Amazon Web Services has said the tool drove nearly $12 billion in incremental sales last year. Jassy said on the earnings call that third-party agents weren't good enough yet -- that they lacked a shopper's history and often couldn't get prices right -- and that people would gravitate to whichever assistant knew them best. That's the opening Amazon is going after with its own AI chatbot and related tools on Amazon.com. "We are aiming to have it be the best shopping assistant anywhere," Jassy said. The strategy reflects one of the ways Amazon is increasingly making money. Advertising is now among its most profitable businesses. J.P. Morgan expects it to bring in about $83 billion in revenue this year and, because the margins are high, to account for roughly a third of the company's operating income. That advertising revenue depends on Amazon getting shoppers to browse its own site rather than handing the decision to an outside chatbot it doesn't control. The big question long-term is whether Amazon can maintain its own role as a primary destination for shoppers and avoid becoming just another selection on a chatbot's shelf.

Amazon's big Prime Day pitch is its AI assistant. Is it working?

It's a big moment for Amazon. The online retail giant's misnamed Prime Day, which actually stretches across half a week and ends June 26, is in full swing, with blockbuster discounts on a vast swathe of products. But rather than leaving shoppers to click, compare, and hunt for deals on their own, Amazon is pushing its AI tools hard. Videos touting the ability to integrate Alexa, or Rufus, Amazon's AI text chatbot, into the shopping experience are a new addition this year. It's a sign of how the company sees the future of online retail, and the strategy is backed up by broader consumer trends. More than half of U.S. shoppers are now willing to let AI handle the entire shopping process, including the final purchase, once their preferences are set, according to payments platform Adyen. Early signs suggest that Amazon's use of AI, and shoppers' own use of AI, has been effective. Adobe, which tracks online shopping behavior during Prime Day and other events, says traffic from AI sources has converted into sales 50.7% better than traffic from non-AI sources. Shoppers who arrived at Amazon from AI sources also spent nearly half as long on the site as other visitors.

This Prime Day, Retailers Make a Bid for the Bots

Amazon's marquee sales event is unfolding in a new environment where A.I. agents research products, compare options and complete transactions. Prime Day may offer the clearest picture yet of how agentic commerce is reshaping retail. Amazon Prime Day has always been a massive opportunity for third-party retailers, which already account for more than 60 percent of the e-commerce giant's sales. The annual influx of shoppers, many of whom have been waiting months to buy wishlist items, offers smaller brands a unique chance to directly compete with household names and capture and convert a wave of new customers. But this year, it's also a stress test against a rapidly emerging technology: A.I. shopping agents. Sign Up For Our Daily Newsletter Sign Up Thank you for signing up! By clicking submit, you agree to our <a href="http://observermedia.com/terms">terms of service</a> and acknowledge we may use your information to send you emails, product samples, and promotions on this website and other properties. You can opt out anytime. See all of our newsletters A.I. agents have become mainstream consumer tools -- 60 percent of shoppers say they're likely to use A.I. to make purchases this year. While the shift is ongoing, Prime Day volume will reveal new insights into how agents are influencing product discovery and purchasing behavior at scale. A bellwether for the future of e-commerce This year's Prime Day transcends the typical hype of a major sales event because it is unfolding in an all-new ecosystem. Artificial intelligence creates an autonomous middleman layer between consumers and retailers that many brands are still racing to accommodate. Zero-click searches have completely disrupted traditional discoverability and ranking tactics. Localization and specificity are superseding competitive keywords. Online stores must be designed not for simple product searches, but for detail-rich conversations. Promotions alone are not enough to stand out. With A.I. agents acting as users' trusted personal shoppers, the new code of online commerce has become hyper-personalization. Brands must be obsessed with their customers, so much so that they can anticipate the bot-human conversations that will trigger a recommendation of their product. The most successful brands will be those that have plugged into these interactions and optimized their storefront to match. The sales event will also assess the technical barriers that have previously hindered the widespread adoption of agentic commerce. Even as A.I. agents become more advanced and platforms adapt to let them transact, A.I. shopping agents are still running into digital walls, such as security protocols and payment verifications, contributing to lower sales across e-commerce vendors. Expect more brands to lower their infrastructural guardrails this Prime Day -- or else, see hits to their performance. As Prime Day kicks off a period of peak e-commerce activity, the agentic handprint will be more visible than ever before. Where and how it appears will shift the blueprint for e-commerce strategy on a global scale. Mapping Prime Day's agentic commerce pressure points This year's Prime Day, already set to break sales records, will provide retailers with robust data to help them understand how agentic A.I. is impacting their businesses. Retailers should monitor the impact of agent-driven shopping commerce on multiple e-commerce touchpoints and behavior indicators. Product discovery is a crucial front: today, most consumers consider A.I.-powered search answers their top digital source when making purchasing decisions. If brands aren't visible here, they'll struggle to reach customers and earn their trust. While reconfiguring their storefronts for A.I. search is already a top priority for retail brands, Prime Day volume will unearth their biggest gaps, but also their biggest opportunities. Their most popular and most-viewed products are more than likely the ones that are showing up in A.I. overviews. Agents take everything a step further: instead of simply providing answers, they conduct the research, deliver recommendations, compile carts and then make purchases. The intelligent recommendation process produces a narrow, personalized set of product options, rather than the broad list of links from a search engine query. The consumer spends less time browsing, making it less likely that they'll stumble across another product on an additional search. Time to decision shrinks. Offering clarity and removing friction become paramount. Retailers are also testing their online reputation against agents. In addition to scanning product briefs, A.I. search tools compile insights from purchase reviews and consumer forums like Reddit. A rise in agentic commerce will show which brands control their online narrative, which are hurting and which are still fighting to be seen. Moreover, this Prime Day is likely to reveal how agents are impacting certain product categories over others. Unlike big holiday shopping events, Prime Day is a moment when many consumers are shopping for themselves and their households. This will no doubt influence how they prompt agents and chatbots to recommend products to them. Instead of asking an agent, "find me gifts under $40 for a mom who loves the outdoors," they might instead prompt, "find me the best deal on suntan lotion with all-natural ingredients for women in their 40s with sensitive skin." Naturally, the impacts of agentic commerce will be most striking in the top-selling categories. Whether retailers are prepared for agent-executed payments could also play a hand in their Prime Day success. Agent-initiated payments are becoming increasingly common as developers find ways around security blocks. Amazon itself has embraced the future of agentic payments with the launch of Amazon Bedrock AgentCore, a tool that helps developers build agents that can successfully interact with online payment systems. Shopify, PayPal and other ecosystem players are likewise investing in ways to accommodate autonomous transactions. Looking past Prime Day, fostering customer loyalty will require securing agent loyalty first. A.I. tools learn continuously from their users' prompts and behaviors. Making a purchase recommended by an agent is positive reinforcement to keep recommending that retailer and similar sites. As A.I. agents evolve to remember and cater to users' favorites, becoming the favorite retailer for both the consumer and the bot will be the new brand imperative. After all, consumers have exceedingly high expectations from these tools: four in five won't return to a brand they discovered through A.I. if they have a poor experience. Adapting to the "new normal" Agentic bot usage now accounts for more than half of all internet traffic. The brands that "win" on Prime Day will be those that have designed their product listings and storefront for what agents prioritize: price, availability, delivery speed and user reviews and recommendations. The value of these insights will extend beyond their Amazon listings. While Amazon is the go-to for many independent retailers, it is not always the one that will best support them. As agentic A.I. becomes more commonplace, retailers will likely need to diversify their channels to remain resilient. This may mean launching on new e-commerce marketplaces like Walmart, as well as seriously considering global expansion. Expansion into international markets is already top of mind for many retailers facing the impact of tariffs, but it offers more than short-term relief, helping them reach more customers, steadily grow their margins and make their brands more recognizable to agents online. Prime Day performance can indicate where brands have the biggest opportunity in a new market. As A.I. continues to disrupt the status quo, having a broader customer base is a form of brand insurance. Meeting A.I. with A.I. Experts predict that 15 percent to 25 percent of all commerce will flow through agentic channels by 2030. As customers embrace automated shopping experiences, retailers will need to match them with automated campaign management and A.I.-powered marketplace infrastructure. For retailers already employing A.I.-enabled e-commerce, Prime Day will show the power of these investments in action. Consider automatic price adjustment. If brands are using A.I. to set the most competitive prices in each market they serve, they're more likely to be discovered by agents that are scouring for the best deals. This is not to say that stores should completely ignore customers in favor of bots. The challenge is to win both the agent's and the customer's loyalty. Even if customers cede the entire shopping process to an agent and never land on a storefront, customer experience is still a retailer's primary differentiator. How brands engage with customers from purchase to delivery and beyond will cement their place as a preferred retailer. Data analytics and A.I.-generated insights will help retailers to understand their customers and deliver personalized recommendations across communications channels, leading to greater customer satisfaction. This Prime Day will reinforce one fundamental truth: agility is the foundation of a successful e-commerce business. Real-time adaptation to customer needs, emerging technologies and market constraints will determine which retailers survive in the age of agentic commerce.

Prime Day Is AI Commerce's Biggest Stress Test Yet | PYMNTS.com

According to a Monday (June 22) Reuters report, Bank of America expects the event to generate $21.6 billion in sales, up just 5% from 2025 -- leaving Amazon little room for the technology to underperform. Amazon moved the event from July and is leaning harder into selling groceries, household goods, travel items and back-to-school purchases. eMarketer expects Amazon to capture more than 60% of sales during the event, according to Reuters, even as Walmart and Target run competing promotions. Alexa for Shopping changes where the buying decision starts. Instead of asking shoppers to scroll through product pages and compare deals themselves, Amazon can use their shopping history and stated preferences to narrow the options before they reach the cart. Amazon said in a June 16 post that the tool can build a personalized Prime Day Deals Guide, explain why each item was selected and send alerts when a matching deal appears. Shoppers can also check price history, set a target price and let Alexa complete the purchase when that price is reached. That puts Amazon's AI inside discovery, comparison and checkout. PYMNTS Intelligence found that 47% of online shoppers used AI during their latest purchase. ChatGPT's share as a product research tool rose from 2% to 30% in two years, the data shows. Retailers now have to compete for the recommendation before a shopper reaches a product page. Prime Day gives Amazon a closed-loop test. The company owns the product data, customer history, pricing, checkout and fulfillment. It can see whether an AI recommendation ends in a purchase. Amazon Uses Alexa to Squeeze More Spending From Existing Prime Members Prime Day has long helped Amazon add Prime members and train them to spend more often. That membership funnel is getting harder to expand in the United States. Last year, Amazon added 3.9 million members in the three weeks before Prime Day 2025, down 185,000 from the prior year and about 193,000 below its goal, according to Reuters. However, the company brought in 1.6 million U.S. Prime members during the last year's event, beating its internal target. Amazon is also widening the purchases Prime Day is built to capture. The company said in its post that this year's event includes deals on pantry goods, pet supplies and household products alongside electronics. Those categories can support repeat orders instead of one large purchase. Prime Day Tests AI Recommendations and Auto-Buy at Scale The commercial case depends on whether the tools work under live retail conditions. Prime Day compresses millions of deals, frequent price changes and time-sensitive buying into a 96-hour window. The annual event is a stress test for AI-assisted commerce at scale. Millions of shoppers making time-sensitive decisions simultaneously is a real load, and the auto-buy feature concentrates the risk: once a shopper grants permission, the system can charge their default payment method and ship to their address the moment a tracked item hits a target price -- no additional confirmation required. For that to work, alerts have to arrive while inventory is still available, and the system has to honor the price and product rules each shopper sets. The last step carries a higher bar than product discovery. PYMNTS reported last week that consumers are more comfortable using AI for recommendations and comparison shopping than for payments and other final decisions. Shoppers still want more control when software moves from advice to spending. Amazon's design keeps several approval levels in place. Shoppers can use Alexa to build a guide, watch a product or authorize an automatic purchase at a set price. Prime Day will put all three uses into the same sales event. In May, Amazon Web Services announced the AWS Agentic Shopping Assistant, built on the same underlying technology as Alexa for Shopping and designed to let third-party retailers deploy comparable tools on their own sites. That move reframes what Prime Day is actually testing. If Alexa for Shopping performs -- if AI recommendations convert, price alerts trigger purchases and auto-buy runs without errors -- Amazon will have a proof of concept it can sell to every retailer that runs on AWS. The real stakes aren't based on Prime Day's success itself. Instead, they hinge on who controls the infrastructure layer of AI-assisted commerce once the event ends.