Generative AI Shopping Assistants and the Future of Google’s Ad Business

The Rise of AI Shopping Assistants: A 1,200% Surge

Consumers are embracing generative AI as the new shopping concierge. Adobe’s latest March 2025 report reveals an astonishing 1,200% year-over-year spike in traffic from generative AI tools to U.S. retail websites. In the 2024 holiday season alone, visits driven by AI assistants jumped 1,300% compared to the prior year (and nearly 20× on Cyber Monday year-over-year). While starting from a small base, this channel’s growth has doubled every two months since September 2024, an exponential trajectory that has caught the attention of marketing leaders.

Not only are more shoppers arriving via AI recommendations, they’re also behaving differently when they land on sites. According to Adobe Analytics (tracking 1 trillion+ retail site visits), visitors coming from generative AI stay 8% longer on-site, view 12% more pages per visit, and bounce (leave immediately) 23% less often than those arriving from traditional channels. In simple terms, an AI-referred shopper might browse, say, 9 pages instead of 8, and is far less likely to abandon the site quickly. This higher engagement suggests that by the time an AI-assisted consumer clicks through, they are better informed and more ready to explore products. Indeed, 92% of shoppers who’ve used AI assistants report it improved their experience often because the AI quickly filters options to their personal needs. Shoppers describe AI chat as a way to “shorten the time required” to get personalized answers.

Critically, this shift isn’t just a holiday fluke – it marks a broader change in consumer behavior. 39% of U.S. consumers have now used generative AI for online shopping, and 53% plan to in 2025. That implies by year-end, a majority of digital shoppers could be routinely consulting an AI assistant. Their reasons vary: 55% use AI to research products, 47% to get product recommendations, 43% to seek deals or discounts, and 35% for gift ideas or finding unique items. In practice, a consumer might ask an AI, “What’s the best espresso machine under $500?” and get a tailored, curated answer in seconds, rather than combing through multiple websites. Others might say, “I need gift ideas for a 5-year-old who loves science,” and receive personalized suggestions. These use cases – from deal-hunting to compiling shopping lists – underscore that AI tools are becoming a front end for product discovery, guiding consumers before they ever visit a retailer’s site.

Even so, marketing leaders should note that AI-sourced traffic remains modest in absolute terms – still a single-digit percentage of total e-commerce visits by Adobe’s estimate. Traditional channels like paid search, email, and direct traffic continue to dominate volumes. But the growth rate and trajectory of AI referrals are what make this trend disruptive. If something is doubling every two months, you ignore it at your peril. 58% of consumers now say generative AI tools (like ChatGPT) have already replaced search engines as their go-to source for product and service recommendations, according to a Capgemini survey. This signals a fundamental shift in where people start their shopping journey – a shift that Chief Marketing Officers (CMOs) and agency leads must factor into strategy immediately.

Beyond retail, Adobe’s data shows this behavior extends to other industries. For example, travel sites saw a 1,700% jump in AI-driven visits (Jul 2024 to Feb 2025) as consumers turn to chatbots for trip planning. Even banking websites saw AI-sourced traffic spike 1,200% in that period as people ask AI for financial guidance. In each case, AI referrals exhibit lower bounce rates and higher time-on-site – e.g. a 45% lower bounce in travel and 45% longer visit duration in banking. The pattern is clear: AI assistants aren’t just sending more traffic, they’re sending more qualified, high-intent traffic across verticals. Consumers arrive having done their homework via AI, be it narrowing down the best TV for their budget or understanding the differences between two credit cards. Increased confidence leads them to dig deeper on sites once they click through.

However, conversion to purchase hasn’t fully caught up yet. Adobe reports that visitors coming from generative AI are 9% less likely to convert (make a purchase) compared to other traffic. The encouraging news is this gap is closing fast – back in July 2024, AI traffic was 43% less likely to convert, so the gap narrowed from 43% to just 9%‍

In numeric terms, if historically only ~60 out of 100 AI-referred shoppers bought something (versus 100/100 other shoppers), now it’s ~91 out of 100. This rapid improvement suggests growing trust and intent to purchase directly from AI recommendations. Early on, people treated AI assistants mainly for research and then bought later via traditional means; now they increasingly complete transactions immediately after an AI chat. We’re essentially watching the funnel compress: the AI guides the consumer through consideration and closer to action in one go. As generative AI integrations improve (and perhaps begin to facilitate transactions directly, as we’ll discuss), we can expect conversion rates from AI referrals to rival, or even exceed, other channels.

Another nuance for marketing teams: AI-driven shopping appears to be happening largely on desktop devices, at least for now. A striking 86% of AI-sourced traffic came from desktop computers, vs. just 14% on mobile, in Nov 2024–Feb 2025. This is essentially the opposite of overall e-commerce, where only ~34% of visits are desktop.

It seems consumers find it easier to chat with AI on a larger screen – typing detailed prompts or reading long answers is more comfortable on desktop. This could change as mobile chat UIs improve or voice interfaces rise (more on voice later), but it’s a reminder that the AI shopping surge is changing the device mix. Desktop experiences – sometimes neglected in recent years – may need a second look, ensuring your website (or client’s site) delivers rich information and easy purchasing for those high-intent desktop users coming from AI assistants.

Lastly, not all product categories benefit equally from AI-assisted shopping. Adobe’s analytics found conversion rates from AI traffic are highest in electronics and jewelry, but lower in apparel, home goods, and grocery. This makes intuitive sense. For complex, spec-driven products like a 4K TV or a laptop (or high-consideration items like jewelry), an AI assistant can parse specs, reviews, and user needs (e.g. “best 65-inch TV for bright room under $1K”) to narrow choices – effectively acting as a knowledgeable salesperson. Shoppers in these categories appear ready to buy once the AI helps pinpoint the right product, leading to higher conversion. In contrast, for apparel or home décor, style and personal taste dominate; an AI can recommend based on description and reviews, but customers may still want to see and feel options (or trust their own eye for style). And groceries are often habitual, list-driven purchases where an AI’s role might be limited (aside from recipe suggestions). For CMOs, this means the impact of AI will vary by vertical – electronics retailers might already see sales lifts from AI referrals, while fashion brands may find AI more useful for inspiration than immediate conversion. Marketing strategies should be nuanced accordingly (e.g. ensure your tech specs and reviews are well structured for AI to ingest if you’re in electronics, whereas apparel brands might focus on integrating AI for virtual try-on or style advice).

Meet the New AI Shopping Platforms: Perplexity, Agora, ChatGPT, and Amazon’s AI Suite

Multiple AI-powered platforms are driving this shift in how consumers discover and shop for products. It’s not just one tool – an ecosystem of generative AI shopping assistants has emerged, each with different strengths. Marketing leaders need to understand these platforms because they are the new gateways for product discovery. Let’s profile the major players and how consumers are using them:

Perplexity.ai – From Answer Engine to Shopping Concierge

Originally known as an AI Q&A search engine, Perplexity has rapidly evolved into a one-stop shopping assistant. In late 2024, Perplexity launched “Perplexity Shopping” features that allow users to search for products via conversational prompts, see summarized results, and even purchase items without leaving the chat interface. For example, a user might ask Perplexity, “Find me a durable backpack for under $150 suitable for airline travel.” Perplexity will return a curated list of backpacks, each with an AI-written summary of features and reviews, along with price and product image.

Crucially, Perplexity introduced a “Buy with Pro” button for its paid subscribers, enabling one-click checkout directly within the chat. If a recommended product supports this feature, a Pro user can literally click “Buy” and Perplexity will auto-fill their saved shipping and payment info to place the order. In essence, Perplexity is trying to remove all friction between research and purchase – you ask, it finds, you buy, all in one continuous flow. For products not (yet) supported by direct checkout, Perplexity simply redirects the user to the merchant’s site but the intent is clear: keep users within the AI assistant for the entire journey when possible.

From a data perspective, Perplexity has built a sizable product catalog by indexing retail data. It has integrated millions of products from large retailers like Amazon and Best Buy, and tens of thousands of smaller Shopify/WooCommerce stores. In fact, the team describes it as “a decentralized Amazon” – an AI-powered search across many stores with a universal cart. This gives Perplexity breadth in results: it might show you a niche product from a small boutique alongside mainstream options. For marketers, this means your product could be recommended even if it’s not on Amazon, as long as Perplexity has indexed your e-commerce site. It’s another reason to ensure your product feeds or site SEO are friendly to AI crawlers.

Interestingly, Perplexity’s leadership has stated they want to avoid creating a new SEO arms race on their platform. “Perplexity’s entire way of operating is so that SEO doesn’t emerge in this new category,” said the company’s Chief Business Officer. In other words, they intend their AI to pick the objectively best or most relevant products based on content and user needs, rather than being gamed by keywords. Instead of classic SEO, the advice to brands is to “build the best product and have that be reflected in the reviews and what others say about it, and then it will naturally rise to the top”.

This philosophy means authentic customer reviews and third-party mentions (e.g. on forums or Reddit) can heavily influence what the AI recommends. Indeed, some savvy brands are already adapting – for instance, posting more on Reddit to ensure their products are mentioned in community Q&A, since OpenAI’s models (which power many AI assistants) ingest Reddit data. We’re essentially seeing the birth of “AI Recommendation Optimization”, distinct from traditional SEO.

In terms of monetization, Perplexity is currently not taking a cut of purchases through its platform. There are no commissions or kickbacks on those “Buy with Pro” sales as of early 2025 – the focus is on improving user experience and gaining adoption. However, Perplexity has launched a merchant program to give brands insights on search trends and to “increase the chance that Perplexity will recommend their products.” This hints at future revenue models: for example, Perplexity could offer sponsored placements or enhanced visibility to merchants, or eventually take affiliate fees once the user base is large enough. For now, the platform’s growth (and the subscription revenue from Pro users) seems to be the priority over immediate ad dollars. But as we’ll discuss later, this may evolve.

Agora – A Decentralized Marketplace Aggregator

Agora is another emergent player, with a different approach. Agora positions itself as an AI-powered search engine for e-commerce products that primarily aggregates independent online stores. Think of it as an AI shopping mall indexing thousands of Shopify and WooCommerce boutiques. As of late 2024, Agora had indexed over 4 million products from 8,100+ small merchants, and that has grown to 13 million products from 35,000 stores by 2025. Its core proposition: help customers discover unique or niche products from small businesses, with the convenience of a single cart checkout across all these stores

From a consumer perspective, a use case for Agora might be: “I want a handmade leather journal from a small US seller” – the AI can search its index and return options from various artisan shops, something a Google search might struggle to compile easily. Agora’s assistant (charmingly named Athena) lets you chat to refine options, plus you can filter by price, compare products, etc., all with AI help. The one-cart feature means if you pick items from three different indie stores, Agora lets you pay once and handles the rest, simplifying what would normally be separate checkouts.

For marketers representing boutique brands, Agora provides a way to get in front of customers who are explicitly looking beyond Amazon. Importantly, Agora emphasizes it charges no commission to the merchants on sales – a very merchant-friendly stance likely aimed at attracting inventory. This suggests Agora might pursue other monetization (perhaps a subscription for shoppers or a tip-jar model, or later introduction of ads). But their ethos is “support small businesses while making discovery easier for customers”. In a sense, they are turning the long tail of e-commerce into something discoverable via AI.

Early adoption of Agora seems driven by consumers who crave unique products or want to avoid the “Amazon effect.” It’s part of a broader consumer shift valuing niche, artisanal, or local products – but now with AI making it as easy to find a handmade candle from a local shop as it would be to search Amazon for a generic one. Digital agencies should keep an eye on platforms like Agora; while they may not yet command huge traffic, they represent a qualitatively different kind of product search (values-driven, niche-focused) that could grow in parallel to the big ecosystems. Also noteworthy: Agora is making itself integratable into other AI assistants – it has an API and even an open-source connector so that AI agents like ChatGPT or Claude can query Agora’s index. This means even if consumers aren’t directly on Agora’s site or app, the platform’s product data could be feeding answers in other AI applications (for example, an AI personal assistant finding a product might pull results from Agora’s database).

ChatGPT (and GPT-4 Search) – Personal Shopper at Scale

No discussion of generative AI in commerce is complete without OpenAI’s ChatGPT, which kicked off the mainstream AI assistant revolution. ChatGPT itself is a general-purpose conversational AI, not exclusively a shopping tool – but its sheer scale of adoption (over 100 million users within months of launch) means many shoppers have experimented with it for product research. In fact, when consumers in surveys refer to using “AI” for shopping, ChatGPT is often the first thing they try (alongside maybe Bing’s chat or others).

How are people using ChatGPT for shopping? Often in a research and brainstorming capacity. For instance, a user might prompt: “I need a gift for my father who loves gardening and tech – any creative ideas?” ChatGPT can generate a list of suggestions (e.g. smart plant sensors, advanced gardening tools, etc.) complete with explanations. Or a savvy consumer might use it to compare product specs: “Compare the top 3 noise-cancelling headphones under $300 and give pros/cons of each.” In seconds, the AI can synthesize review content and spec sheets into a handy summary. This deep research capability is why 55% of AI-shopping users cite “conducting research” as a top use. ChatGPT, especially with GPT-4, excels at digesting large amounts of info (like multiple product reviews) and presenting a concise answer. It’s like having a personal product analyst on call – something that appeals greatly to busy consumers.

One limitation: ChatGPT’s default knowledge cutoff means it doesn’t innately know about the very latest products or real-time prices. However, OpenAI has introduced web browsing plugins and integrations that many users leverage. With web access enabled (or via plugins like the former Klarna shopping plugin or Instacart plugin), ChatGPT can pull current product data, availability, and pricing. For example, using a shopping plugin, a user could ask, “Find me the best price for a Nikon D3500 camera”, and ChatGPT might return live pricing from various retailers. Even without a plugin, some users simply copy-paste info – but more seamlessly, ChatGPT is integrated into Bing’s search engine for those using the Bing Chat interface, meaning it can search the web live.

Speaking of Bing Chat (Microsoft’s GPT-4 powered search assistant) – it functions similarly for end-users and likely accounts for a share of that AI-driven traffic spike. Bing’s AI will often answer a product query with a conversational summary and direct links to suggested products (sometimes affiliate links or ads). Microsoft reported early on that they saw increased engagement with their Bing Shopping features via the chat interface, and they’ve been experimenting with ads embedded in chat responses. For example, Bing Chat might respond with *“The Sony WH-1000XM5 headphones have top-tier noise cancellation and 30-hour battery life, making them a great choice and alongside this answer, display an ad unit or shopping card for that Sony model.

So, whether through ChatGPT’s own interface or Bing’s AI search, OpenAI’s tech is steering a lot of product discovery. The key for marketers is that these answers tend to favor concise, content-rich information. If your product has a lot of well-documented benefits and positive discussions online, ChatGPT will surface that. If not, it might ignore your brand altogether. There’s no “paid” way to get into ChatGPT’s responses yet (we’ll explore future monetization ideas later), so the playing field is a mix of brand authority, user-generated content, and sometimes randomness in how the AI was trained. It’s a new kind of challenge: instead of optimizing for a search algorithm, you’re almost optimizing for an AI’s understanding. Some companies are already creating AI-oriented FAQs and content on their sites, knowing that “by 2026, over a third of online content may be created to work with AI-powered search engines”. This proactive content strategy can help ensure an AI like ChatGPT gets accurate, rich info about your products.

One also cannot overlook the user trust ChatGPT has built in a short time. It speaks in a human-like, authoritative tone which many users find convenient (even if it sometimes errs). That trust is evidenced by the Capgemini stat earlier – a majority of users in that survey said AI assistants have become their go-to over search for recommendations. For straightforward questions like “What’s a good budget smartphone?”, a chunk of your audience might not be Googling at all – they’re getting one or two answers from ChatGPT. And while ChatGPT won’t complete the purchase (it will likely mention a few models and perhaps where to buy them), it has massive influence on the consideration set. If your product isn’t among those one or two mentioned, it effectively doesn’t exist to that consumer. This is raising the stakes for AI-era content marketing – brands need to be part of the knowledge base that AIs draw from (through press, reviews, forums, etc.) since there’s no straightforward “ad buy” to guarantee placement here yet.

Amazon’s AI Arsenal: Rufus, “Interests,” Health AI, and Alexa+

While independent AI assistants rise, Amazon – the king of product search – is infusing generative AI across its platform to maintain its dominance. Amazon knows millions start their shopping on its site/app already, and generative AI is a double-edged sword: it could either divert those searches elsewhere or supercharge Amazon’s own experience to keep customers loyal. Thus, Amazon has rolled out a suite of AI shopping tools:

• Rufus – the AI Shopping Assistant in the Amazon app: First tested in mid-2024 and now available to all U.S. customers (as of Q4 2024), Amazon’s Rufus is a conversational agent integrated into the Amazon mobile app Customers can tap a button on the app and literally chat with Amazon’s catalog. For example, one might ask, “I need a gift for a college student who loves cooking in the dorm – what do you suggest?” Rufus will respond with a short list of product recommendations (pulled from Amazon’s vast inventory) along with reasoning: it might say “How about an Instant Pot Mini? It’s compact and multi-functional, perfect for a dorm.” Under the hood, Rufus is drawing on Amazon’s product data, customer reviews, and existing Q&A content to formulate answers. Essentially, it’s synthesizing what previous buyers have said and what’s in the description to answer user questions in plain English. Early user feedback indicated it’s like asking a very well-read fellow shopper for advice.
  ‍
• Rufus shines for questions that aren’t easily answered by filters. For instance, “Is this TV good for gaming in a bright room?” – traditionally, a user would have to read reviews to find mentions of gaming and brightness. Rufus can gather those snippets (e.g. “multiple customers mention low input lag and high brightness on the Samsung 55” TV, so it’s likely great for gaming even in bright rooms”) and give a single answer. From a business standpoint, this keeps the user engaged within Amazon rather than bouncing out to Google to research. It’s effectively an in-house AI-powered sales associate.
  ‍
• Importantly, Amazon can monetize Rufus indirectly through increased conversion. If Rufus helps customers find the right product faster, they’re more likely to buy (and buy from Amazon). And Amazon could later introduce sponsored results within Rufus’s answers (e.g. have the first recommendation be from a brand that paid, labeled as “Sponsored”). For now, Amazon has been cautious to make Rufus helpful and win trust; overt ads in the answers haven’t been widely observed yet, but given Amazon’s ad business, it’s certainly possible down the line.
  ‍
• “Interests” – Personalized AI Product Discovery: Launched in March 2025, Amazon Interests is a feature that lets shoppers describe a passion or hobby, and then continuously gets AI-curated product recommendations over time. Essentially, you feed it a prompt like “the latest pickleball gear and accessories” or “vintage-style jazz posters”, and Amazon’s AI will show you relevant items and notify you when new products matching that interest arrive. It’s an AI-driven twist on product discovery and wishlist combined. Rather than you searching periodically, the AI proactively surfaces new inventory that fits your described interest. This drives ongoing engagement – a marketer might see this as creating a persistent personalized storefront for each user.
  ‍
• Amazon Interests is only available to a subset of users (rolling out gradually), but it signals Amazon’s intention to lock in users’ attention with AI. If a customer tells Amazon their specific interests, Amazon can essentially algorithmically generate a curated feed – and of course, it can mix in sponsored products in that feed relevant to those interests as a new ad channel. For example, if “pickleball gear” is your interest, a new paddle brand could pay Amazon to feature their paddle when the AI feed updates. The monetization potential here is high, but it hinges on Amazon delivering genuine value so users actually adopt the feature.
  ‍
• Health AI (Wellness Chatbot): Amazon has also been testing a generative AI health and wellness assistant in its app. This tool allows users to ask health-related questions (e.g. “What vitamin supplements can help with better sleep?” or “How to alleviate lower back pain?”), to which the AI gives conversational advice and product suggestions (with the appropriate disclaimers). It’s a natural extension: health queries often lead to product recommendations (vitamins, devices, books, etc.), so Amazon can capture those inquiries directly. By addressing these through an AI, Amazon again keeps users in-house instead of losing them to WebMD or Google. For marketers in healthcare or CPG, this means your product might be recommended by Amazon’s AI as part of a health solution (if you have good reviews, ratings, and relevant use cases).
  ‍
• Imagine a scenario: A user asks, “How do I improve my indoor air quality?” Amazon’s Health AI might respond with some general tips and then suggest a couple of air purifiers available on Amazon, citing features and customer ratings – essentially combining content and commerce. This is a potent mix of content marketing and direct selling, automated through AI. It’s still early (limited user testing), but it demonstrates how Amazon can leverage AI to enter domains like health advice which traditionally were outside a retail app’s scope.
  ‍
• Alexa+ (Next-Gen Voice Assistant): Amazon’s voice assistant Alexa is getting a generative AI upgrade, unofficially dubbed “Alexa+” by tech observers. The new Alexa (rolling out to Echo devices and Alexa app) is described as “smarter, more conversational, more capable” thanks to a large language model brain. For consumers, this means hands-free shopping conversations. One can say, “Alexa, find me a highly-rated espresso machine under $500 and schedule it for delivery”, and Alexa+ might talk you through options, answer follow-up questions (pulled from reviews/specs via generative AI), and place the order.
  ‍
• Voice commerce has long been an ambition (remember the hype around voice ordering circa 2018?), but adoption was tepid because voice interfaces weren’t smart enough. Generative AI changes that by making Alexa able to handle complex queries or preferences in a single utterance. Alexa+ can remember context, compare options, and clarify – basically acting like a personal shopping assistant you can speak with casually. From a strategic perspective, this keeps Amazon deeply entrenched in consumers’ lives. If someone gets used to simply asking Alexa for any shopping need, that bypasses Google completely. It’s the ultimate walled garden scenario: Amazon not only is the store, but also the search and recommendation engine for the customer, via an AI personality that people might form a habit around.
  ‍
• For brands, Alexa+ could become an advertising channel as well – e.g. “Alexa, I need new batteries” might default to Amazon Basics unless a brand has paid for placement or has high loyalty. We might see voice ad sponsorships where Alexa suggests a sponsored brand in its answer. Again, Amazon will tread carefully to not undermine user trust, but one can envision a future where bidding for the “AI Suggestion Slot” on voice becomes akin to bidding on keywords today.

In summary, Amazon’s multi-pronged AI strategy (Rufus Q&A, Interests feed, Health advisor, Alexa voice AI) is about meeting the customer wherever they need guidance and keeping them within Amazon’s ecosystem. The convenience is undeniable: over 60% of online shoppers already start on Amazon for product searches historically, and these AI features aim to capture the remaining scenarios where someone might have gone to Google or another site for advice.

From a marketing leader’s lens, Amazon’s moves mean two things: (1) Amazon will remain a critical channel, possibly even more so as these AI features drive engagement and sales (so continuing to invest in Amazon SEO and Amazon Ads is wise), and (2) the nature of Amazon optimization will evolve. It won’t be just about bidding on keywords; it will be about ensuring your product content (descriptions, titles, customer Q&A, reviews) is rich and relevant so that Amazon’s AI picks your product when a shopper asks a broad question.

For instance, if you sell a camera tripod and lots of people ask Rufus “what’s a good lightweight travel tripod?”, you’d better have in your content or reviews phrases like “lightweight” and “great for travel” so the AI associates your product with that need. Incremental digital PR like encouraging customers to use the Q&A section or leave detailed reviews can directly influence AI recommendations on Amazon. One Amazon-focused agency, Incrementum Digital, even advises brands to “add text to product listings that corresponds to common questions asked on Rufus” – essentially, anticipate what users will ask the AI and bake the answers into your listing. That is a tangible new tactic in the age of AI commerce.

Google’s Strategic Crossroads: Will AI Assistants Disrupt Search Ad Dominance?

For the past two decades, Google Search has been the gateway through which consumers discover products online – and the toll collector for advertisers via its Ads. Google’s Shopping Ads and paid search listings have been a staple in marketing budgets, driving billions in sales. But the rapid rise of generative AI shopping assistants is challenging Google’s stronghold on product search in a way not seen since the emergence of Amazon itself.

The threat to Google is twofold:

1. Fewer “top-of-funnel” search queries on Google as users shift to AI assistants. If a sizable portion of consumers start asking ChatGPT or Perplexity for product advice first (instead of typing a query into Google), Google simply loses those eyeballs and queries. Fewer queries mean fewer ad impressions served. We already see evidence of this shift: 58% of consumers say AI tools are replacing search engines for them in product recommendation tasks. That is a stunning figure – it suggests more than half of users may often skip Google for certain shopping queries. For Google, every query not asked is a potential ad click lost. Historically, Google has faced competition from Amazon (with estimates that ~50% of product searches start on Amazon). Now, the other ~50% that might have been Google’s turf is getting nibbled away by AI assistants and specialized search like Agora.

Concretely, think of queries like “best budget 4K TV” or “gift ideas for 10-year-old boy”. These are lucrative queries that would trigger multiple ads on Google. If now a user poses that to an AI, Google doesn’t even get a chance to show an ad. Over time, a widespread adoption of AI shopping assistants could erode Google’s query volumes, especially on the high-intent, product-focused searches that advertisers pay premium CPCs for.

2. Even when users stay on Google, the experience is shifting to AI-driven answers with fewer ad opportunities. Google has not stood still – it introduced its own Search Generative Experience (SGE) in 2023 and has been refining it. In SGE, Google shows an AI-generated summary at the top of the search results for certain queries. While Google has begun integrating ads into this AI Overview, the format is different from traditional search results. Early testing showed that in about 27% of cases, the AI answer was shown with no ads next to it, and in 73% cases ads were present. So sometimes the AI result essentially pushed organic content down without an ad, and other times Google did include ads – often in the form of product listing snippets or links above or below the AI box.

From a revenue perspective, Google is cautiously finding ways to inject ads into AI answers. For example, Google announced in late 2024 that it would start displaying Shopping Ads directly within the AI snapshot for commercial queries. In practice, that might look like an AI-generated paragraph recommending a few products, alongside a couple of sponsored product listings with an “Ad” label. Google is essentially trying to marry its ad business with the new AI interface. However, it’s a delicate balance: too many ads could deteriorate the user experience that the AI is supposed to improve. And too few ads (or lower click-through on AI results) could mean less revenue. Early data from the SGE tests indicated that while ads still appear, the number of ads users see may be fewer than in a classic 10-link search result page (especially if the AI gives the answer without need to scroll as much). One study observed that some SGE results had no traditional ads, especially if the AI could answer directly. For advertisers, it raises concern that their paid search spots might not show up as frequently, or might be shunted to less prominent positions.

The strategic implication: Google’s search advertising model is entering a paradigm of diminishing returns on certain query types. If AI assistants keep users within their ecosystems, Google loses queries. If Google’s own AI answers reduce the need to click (the classic “zero-click search” problem, now exacerbated by AI summarizing answers), then even when queries happen on Google, fewer clicks on results (including ads) may occur. It’s notable that nearly a quarter of digital marketing budgets are spent on search. Marketers are thus paying close attention – if those dollars start yielding less traffic or conversions because the search landscape changed, budgets will shift elsewhere.

Google of course is not ceding ground easily. We can expect Google to double-down on making its Shopping products more AI-enhanced and interactive. For instance, Google has been working on visual search and AR try-ons, and integrating those into an AI chat could be a differentiator. It’s also possible Google will leverage its Android ecosystem to push its AI shopping help (imagine an Assistant on your phone that can proactively give shopping suggestions – Google might preempt others by being baked into devices).

However, one area Google lags is first-party commerce data. Amazon’s AI has direct access to conversion and inventory data on its platform, whereas Google relies on indexing websites. If more transactions move to environments like Amazon or even Perplexity (with direct checkout), Google could lose insight into consumer purchase behavior. That data loss further weakens Google’s ad targeting over time.

From Google’s revenue perspective, some analysts have tried to quantify the risk. Morgan Stanley, for example, estimated that if half of Google’s searches were handled by an AI that gave longer answers, it could increase Google’s costs by $6 billion annually due to the computing expense (AI answers use more processing power than standard search).

At the same time, if those AI answers reduce ad volume, it’s a double hit: higher costs, lower revenue. It’s no surprise Alphabet’s stock trembled whenever news broke of advances in ChatGPT or that aforementioned Adobe stat of 1,200% growth – investors are calculating the potential impact on Google’s $162 billion (in 2022) search ads business.

Will Google’s Shopping Ads revenue be cannibalized? In the short term, Google is trying to adapt by blending ads into new formats. We might see new ad products like sponsored chat responses or product carousels within the AI snapshot. Google could also emphasize formats that AI can’t easily replace – for instance, video ads on YouTube for shopping inspiration or discovery ads that appear in Gmail/Discovery feed. But make no mistake, if user behavior shifts significantly, marketers will follow the eyeballs (and clicks).

Already, brands are noticing that the old playbook for search optimization is getting an AI rewrite. An eMarketer analysis notes that retailers are scrambling to find ways to “optimize for AI search” because traditional SEO and search ads may become less effective as AI assistants insert themselves in between brands and consumers. In the long run, if AI assistants (be it Amazon’s, OpenAI’s, or others) become powerful intermediaries, they could even command their own ad ecosystems, forcing marketers to pay them for placement the way we do Google today. We’ll explore that in the next section.

For now, what should Google-focused advertisers do? Keep a close watch on performance metrics for search campaigns, especially on product-related queries. If you start seeing impression or click dips that can’t be explained by competition or seasonality, it could be the subtle effect of changing user behavior (e.g., fewer people searching that term on Google at all). Also, marketers should participate in Google’s beta programs (like SGE) and understand how their ads or content appear in those contexts. Google has stated that “search ads will continue to play a critical role” in the new AI-powered search experience, and that ads will be present in AI results as appropriate, but the format and dynamics may shift. Advertisers might need to adapt ad copy – for example, if an AI overview is already summarizing information, maybe the ads need to emphasize a promotion or unique value prop to stand out.

Google’s product search dominance is under pressure. The next 12-24 months will be critical to see if Google manages to keep shoppers within its ecosystem via its own AI or if they bleed out to other assistants. For CMOs, it’s a time to diversify your traffic mix and not rely solely on Google Shopping Ads as the only gateway to consumers. The pie of “search” is being sliced into traditional search, AI assistant referrals, and on-platform search (Amazon, etc.). Winning in this new landscape means rethinking how you allocate budgets across these entry points.

Agency Playbook: Adapting Strategy in the Age of AI Shopping

For digital marketing agencies and in-house marketing teams alike, the rise of AI shopping assistants demands evolution in strategy, media mix, and client education. Here’s how forward-thinking agencies are responding:

1. Embrace a Multi-Channel Mix (Beyond Google Search): It’s time to revisit the media mix model. If previously you allocated, say, 50% of performance budget to Google Search, 30% to social, 20% to Amazon, the new landscape might warrant adding new channels and adjusting weights. Invest more in Amazon and retail media: Given Amazon’s aggressive AI integration (Rufus, etc.), we anticipate even more product search activity staying on Amazon.

Ensure your clients’ Amazon Advertising (sponsored products, brands, etc.) is fully funded and optimized – if Google starts to underperform, shifting some budget to Amazon could capture those conversions. Also, explore Microsoft’s ecosystem: Bing’s chat and AI advancements are driving more usage of Bing. Microsoft Ads (which serve on Bing) now have the opportunity to reach users in the Bing Chat interface. If you haven’t already, allocate test budgets to Bing Shopping Ads or multi-channel campaigns. The competition is lower than Google, and any spillover from ChatGPT users clicking a Bing result could be valuable.

2. Optimize Content for AI Assistants (the new SEO): As discussed, we’re entering an era where instead of optimizing just for Google’s algorithm, we need to optimize for AI consumption. Concretely, this means:

• Ensure your product pages have detailed, quality content (features, specs, usage information) because AI models love rich data.
  ‍
• Add FAQ sections addressing common questions shoppers ask. If people often ask “Is this couch pet-friendly?” and you have that answered on your site, an AI like ChatGPT or Rufus is more likely to include your product in an answer about pet-friendly couches.
  ‍
• Structured data is your friend. Use schema markup for products, reviews, etc., so that AI scrapers or APIs can easily ingest accurate info about your products.
  ‍
• Monitor AI search results: Act like a secret shopper – ask ChatGPT or Perplexity some key queries relevant to your client’s product. See which brands get mentioned. If your brand isn’t showing and a competitor is, analyze why. Do they have more content out there? More press? Perhaps generate some thought leadership or how-to guides that indirectly make your brand more “known” to the AI. For example, a cookware brand might publish a guide to baking (mentioning their own products), which an AI might later pull info from when asked about bakeware.
  ‍
• It may sound strange, but think about “AI PR”. Influence the inputs to these AIs by ensuring your products are talked about in the digital sources AIs train on: reviews sites, forums, YouTube, etc. AIs often draw answers from aggregated opinions – so the more positive, detailed chatter about your product online, the better.

3. Educate Clients with Data and Poise: Managing client expectations is key. Some clients might see dips in Google performance or overall site traffic and panic. It’s the agency’s job to contextualize these shifts as part of a broader trend, not necessarily a failure of execution. Use the data: “Traffic from AI assistants to retail sites is up 12x year-on-year. And those visitors behave differently,” you might explain.

Show how engagement metrics from AI referrals are strong (lower bounce, more pages) to reposition the conversation around quality of traffic, not just quantity. If conversions via Google are soft, discuss how user journeys are fragmenting – a customer might consult ChatGPT, then go directly to a brand site or Amazon to buy, bypassing Google ads. Thus, last-click attribution might miss the AI influence. You may need to help clients update their attribution models or at least acknowledge these new touchpoints.

Also, reassure that being proactive is the best response. Present an action plan (much like this list) to clients highlighting how you’ll adapt: e.g., “We are reallocating 15% of your search budget to Amazon and Bing where we see growing AI-driven activity. We’re also implementing an AI-focused content refresh on your top product pages so AI assistants will pick up the latest info. We’ll report back on changes in traffic composition.” This positions your agency as on top of industry changes and turning them to the client’s advantage.

4. Experiment with New Ad Formats and AI Partnerships: The marketing leaders who thrive in this environment will be those who experiment early. Keep an eye out for advertising opportunities on AI platforms. Some are already emerging – for instance, Microsoft has quietly started offering sponsored slots in Bing Chat results. Amazon might let select brands beta test sponsored responses from Rufus (imagine paying to ensure Rufus mentions your new product launch for relevant queries – likely an invite-only trial at first). Even startups like Perplexity or Agora could roll out promoted product listings in their interfaces down the line.

Volunteer your clients for beta programs or alphas. Not only might you get cheaper clicks before auctions get competitive, but you’ll learn what works in these novel formats. The first-mover advantage is real: if you’re the first brand in your sector to master advertising on, say, an AI shopping app, you could capture disproportionate share of that audience.

Additionally, consider partnerships with AI platforms. For example, some forward-thinking agencies are in talks to feed their product catalogs to AI systems directly. If you represent a large catalog, perhaps you can provide a product data feed to an AI assistant company so that it guarantees awareness of your products. This is akin to ensuring you’re in the “knowledge graph” of the AI world.

Internally, also use AI to boost your own operations – this is tangential, but worth noting: agencies are using generative AI to automate ad copy creation, A/B test ideas, and analyze large datasets. If AI is disrupting how consumers shop, it can equally be a tool to improve how we market (e.g. using AI to predict which products an AI might favor – a bit meta, but doable).

5. Rethink Creative and Messaging: In an AI-driven discovery, the context in which your ads or content appear might be different. For instance, if someone is using an AI assistant, by the time they see a brand message, they might already be in research mode for 10-15 minutes with the AI. They might have a quasi-“relationship” with the AI assistant, trusting its recommendations.

How do you piggyback on that trust? One approach is to ensure any messaging aligns with the kind of concise, informative tone that AI uses. If you’re running a paid ad that might appear near an AI answer, consider making the copy more useful and factual (to feel like a continuation of the assistant’s help, rather than a jarring sales pitch). For example, instead of a generic tagline, an ad might say “Top-Rated 4K TV – 5-Year Warranty, under $500,” which is a very factual, appealing snippet that an AI might have said itself.

Also, the creative formats might evolve. We might see interactive ads where users can ask a question to a brand’s bot on the spot. For instance, Google has hinted at “conversational ads” in the future – imagine an ad unit that lets the user chat with a product’s AI rep. Marketers should start thinking about content that can be delivered via conversation, not just static text or images. Training a custom mini chatbot on your product catalog for user engagement could even become part of performance marketing.

6. Double Down on First-Party and CRM Strategies: If AI intermediaries stand between you and consumers, building direct customer relationships becomes even more critical. Encourage clients to invest in their loyalty programs, email/SMS marketing, and community building.

The goal is to have consumers come directly to you or at least seek you out by name. Brand loyalty can override some AI influences – e.g., a user might explicitly ask an AI, “What’s the latest product from Brand X?” if they feel connected to Brand X. Strong brands will have an edge in an AI-driven world, because even an AI can recognize brand reputation and user preference (especially if the user instructs it by brand). So from a broader strategic view, agencies shouldn’t neglect upper-funnel brand campaigns and customer experience, thinking only about the tactical AI search wins. It all ties together.

In essence, agencies must become nimbler and more consultative. This is a moment to shine as strategic partners – navigating clients through uncharted territory with a steady hand on the data and an eye on innovation. Nearly 91% of US ad agencies are already exploring or using generative AI in some capacity, indicating that the industry is collectively moving to adapt. The ones who translate that exploration into concrete strategy shifts (as above) will set themselves – and their clients – apart.

The Future of Monetization: How AI Shopping Assistants Could Make Money

Today, many generative AI shopping experiences are ad-free or in pilot monetization stages, focused on user growth and product-market fit. But the big question is: Where is the money in this long term? If AI assistants become as important as we think, how will they sustain themselves or profit? Let’s put on a speculative yet strategic hat and explore how platforms like Perplexity, ChatGPT, and Amazon’s AI features might monetize these shopping interfaces via paid advertising or other models:

Perplexity & Agora – Affiliate Models and Sponsored Listings: As noted, Perplexity currently doesn’t take a commission on “Buy with Pro” sales. That’s likely a temporary stance to encourage usage. In the future, Perplexity could adopt an affiliate revenue model: for each sale it facilitates (especially if it directs out to a merchant site), it could earn a referral fee. Given it’s aggregating major retailers, it could negotiate affiliate deals (many retailers already have affiliate programs, often giving ~1-5% of the sale). If Perplexity becomes a significant traffic source, expect them to quietly start receiving those fees. They may keep the experience “ad-free” in appearance, monetizing on the back-end via these commissions – a bit like how credit card comparison sites make money per card sign-up but to the user it just looks like rankings.

In addition, sponsored product placements are a logical extension. Perplexity could maintain the integrity of its answers but have, say, one of the recommended products be a paid placement (marked subtly as “promoted”). This is analogous to Google’s search ads vs organic results, but in a single merged list. Given Perplexity’s CBO talked about avoiding SEO gaming, they might prefer an explicit ads model to keep the rest truly organic. For example, if a user asks for “best running shoes”, Perplexity might show 5 options. Perhaps option #3 could be a brand that paid to appear, labeled as sponsored. If done transparently, users may accept it as long as it’s relevant.

Agora, serving the long-tail merchants, might also use sponsorships – maybe a small merchant can pay a fee to get their products highlighted for certain keywords (like an Etsy promoted listing). Or Agora could charge a small transaction fee to consumers or merchants once it has scale, especially since it adds convenience (one cart). However, stating “no commission” suggests they might instead opt for a subscription or premium model (e.g., a Pro version of Agora with perks, or charging for enhanced listing exposure).

ChatGPT (OpenAI) – Native Ads or Subscription-Only? OpenAI’s ChatGPT has a subscription (Plus) for users, which might indicate they lean toward a subscription model over advertising for revenue. Sam Altman (OpenAI’s CEO) has historically been cautious about the idea of turning ChatGPT into an ad-supported platform, as it could bias the assistance. That said, the lure of advertising dollars is huge. Imagine the scenario: ChatGPT gets, say, 1 billion queries a day (just hypothetical). Even if 5% of those are commerce-related, that’s 50 million high-intent queries daily. If OpenAI served an ad or affiliate link on even a fraction, the revenue could be significant.

One potential path is affiliate partnerships similar to what we described for Perplexity. OpenAI could partner with an e-commerce aggregator (maybe they quietly partner with Microsoft’s Bing Shopping or others) so that whenever ChatGPT’s answer includes a product suggestion, it also provides an affiliate link. For instance, if ChatGPT says “I recommend the Nike Air Zoom Pegasus for a reliable running shoe,” it might include a link to purchase it on a partner site. Already, Bing Chat does something akin to this, as it often deep-links to merchant sites and Microsoft likely has affiliate arrangements. OpenAI could do it too, or rely on Bing’s integration (since ChatGPT with browsing currently uses Bing’s API which might already attach referral codes).

Alternatively, OpenAI might allow sponsored knowledge panels – e.g., a company could pay to have an official, verified snippet used by ChatGPT. Picture asking about a product and ChatGPT using the manufacturer’s provided description (with a note). This is speculative, but OpenAI could create a program where brands submit information to ensure accuracy (for a fee). It’s less about an obvious ad and more about data placement.

However, OpenAI also has to consider its broader strategy. They’ve positioned themselves as a research/tech provider and partner with businesses (ChatGPT API, etc.). They might leave consumer-facing monetization (like ads) to partners like Microsoft, and instead earn via API calls. In that case, the monetization of AI shopping might be indirect for OpenAI – e.g., charging Bing and others for using its models that facilitate these searches. But as standalone ChatGPT continues to be used, I suspect we may see at least light-touch monetization for shopping queries by 2025-2026. Perhaps a “Shop Now” button powered by a partnership when ChatGPT lists a product recommendation.

Amazon – Sponsored Answers and Paid Inclusion in AI Features: Amazon’s monetization strategy for its AI tools will likely integrate with its massive advertising machine ($38 billion/year and growing in ad revenue). We should expect that Sponsored Products will find their way into Rufus responses and Interests feeds. For example, if three products fit a query, Amazon could insert a fourth one that is a sponsored listing (clearly labeled). Given users are already accustomed to “Sponsored” tags on Amazon search results, doing similarly in an AI chat likely won’t cause much friction, provided the recommendation is on-point.

Amazon could also offer brands the opportunity to train the AI on their product details for better answers. Perhaps a brand pays for a program where Amazon’s AI is given enhanced data (like assembly videos, detailed specs) about their products, ensuring when someone asks a tricky question, the AI can answer it correctly for that brand’s items. It sounds more like a value-add service than an ad, but it’s monetization – maybe rolled into higher seller fees or a premium listing service.

The “Interests” feature has clear ad potential: if a user has, say, “smart home gadgets” as an interest, relevant brands could sponsor a “featured new product” that shows up in that feed. Since Interests is about notifying users of new products, a brand launching something new in that category might pay to be included in the next update the user sees. It’s almost like retargeting meets content marketing: user says what they want, and you ensure your product finds them.

One must consider user trust: if these AI assistants become too ad-saturated, users might look for alternatives. It’s a fine line. The likely scenario is a hybrid monetization: some subscription or premium upsells (like Perplexity Pro), combined with careful introduction of sponsored content. The goal is to mirror the success of search ads – which, when done well, appear as helpful results while being paid – in the AI context.

Bing & Google – Already Paving the Way: We have a preview in Bing and Google’s moves. Bing Chat has been injecting ads contextually (e.g., if you ask for a product, you might see a small ad with it). Microsoft even reported higher click-through on some of these because the AI would mention a product and the ad was right there to fulfill it. Google’s SGE will likely formalize AI-driven ad formats. For instance, Google could sell “Conversation Ads” where a brand’s message or product is woven into an AI interaction. They’re testing things like AI-organized results pages that cluster info – a sponsored cluster could be a thing (imagine a section “Brand X’s Guide” appearing via AI arrangement, paid by Brand X).

New ad metrics and challenges: If users start interacting via conversation, how do we measure ad performance? It may shift from simple clicks to engagement within the AI. Possibly, voice commerce (with Alexa+) will force new metrics like “share of voice” – did the AI recommend your brand when prompted? We might see the rise of AI Share of Recommendation (AI-SOR) as a KPI: the percentage of times an AI recommends your brand in relevant queries. There might even be bidding for that – e.g., upbid to increase the likelihood of recommendation unless it conflicts with user intent.

We should also watch for monetization beyond ads: AI shopping assistants could charge users in creative ways. Perhaps a premium AI shopping service that guarantees the best deals (e.g., an AI that will automatically apply coupons or negotiate prices for you for a monthly fee). While consumers love “free”, they might pay if the AI demonstrably saves them money. For instance, imagine an AI that monitors price drops for items in your interest list and auto-refunds you the difference (some apps do price protection; an AI could streamline it). The AI could take a small cut of the refund as a fee. These indirect models could emerge as well.

From a marketer’s standpoint, however, advertising and promotional visibility will be the main concern. It’s reasonable to expect that by 2026, we’ll have something like programmatic ad buying for AI assistants. Perhaps an extension of DSPs (Demand-Side Platforms) where one of the inventory options is “AI Assistant Ads” across various partners. The format might not be a banner or search ad as we know it, but a snippet of text or a suggested product card within a chat.

Marketing leaders should start envisioning campaigns in which the target is an AI agent: “How do I get Alexa or ChatGPT to favor our brand when users ask about a problem we solve?” That may involve a mix of classic advertising (if the AI accepts bids) and good old-fashioned brand building so that the AI’s knowledge graph thinks of your brand organically.

One speculative but illustrative example: imagine influencer marketing meets AI. If an AI is trained on content from influencers or experts, having those folks authentically mention or recommend your product could indirectly make the AI more likely to mention it. This is like secondary monetization: you pay an influencer to plug your product (we do that now), but the audience isn’t just their human followers – it’s also the AI models digesting YouTube videos, blog posts, etc. We already saw brands posting on Reddit intentionally to influence AI. These are quasi-paid tactics (time or money invested) to shape AI outputs without paying the AI platform directly.

Finally, consider the scale of the opportunity: Morgan Stanley projects generative AI could drive $1.1 trillion in new revenue by 2028 across various sectors – a lot of that will be advertising and commerce. We’re essentially on the ground floor of an entirely new ad paradigm that could rival search and social ads. It’s an exciting and disruptive time. Companies like Google, Amazon, Microsoft are well positioned to capitalize, but so are nimble startups that can define a new ad experience (think how Google created AdWords; an AI startup might create the equivalent for conversational commerce).

In summary, while today’s AI shopping assistants might not bombard you with ads, commercialization is inevitable. The likely scenario is a confluence of models: affiliate commissions, sponsored answers, data partnerships, and premium user fees. Marketers should anticipate this and be ready to participate in beta monetization programs. The early winners will lock in favorable economics (just as early Google Ads advertisers enjoyed cheap clicks). At the same time, maintaining consumer trust will require relevance and restraint in advertising. Those who strike the right balance – delivering helpful suggestions that also happen to be paid placements – will define the new best practices of AI-era marketing.

Conclusion: Embracing the AI Shopping Revolution

We are witnessing a seismic shift in digital consumer behavior: generative AI shopping assistants are moving from novelty to norm. A year ago, the idea that 1 in 3 shoppers would use an AI chatbot to decide what to buy seemed futuristic – now it’s reality, with nearly 40% already doing so and over half planning to join in. The 1,200% surge in AI-driven traffic to retail sites is a loud signal that consumer expectations are changing. People want instant, personalized, conversation-like experiences in shopping, not just static search results or endless product grids.

For Google, this means adapting or risking erosion of its dominance in product search – the old playbook of “consumers will see our search ads” is no longer a given when the consumer might not search at all, but rather ask an AI that doesn’t show Google ads. For Amazon, it’s an opportunity to deepen its relationship with customers, turning the entire shopping journey – from discovery to purchase – into a seamless, AI-curated experience. For emerging platforms like Perplexity and Agora, it’s open season to invent new paradigms and perhaps become the “next Google” of the AI era (with far less overhead content – it’s all AI synthesized).

Marketing leaders and agency professionals should view this not with fear, but with strategic enthusiasm. The fundamentals of understanding your customer and meeting them where they are remain unchanged – it’s just that “where they are” might soon be talking to an AI in a chat window or via a smart speaker. The companies that thrive will be those who meet customers in these new venues: by ensuring their brand is present and persuasive when an AI is effectively selling on their behalf, and by leveraging the same technologies to create smarter campaigns.

A few closing recommendations for CMOs and digital advertising VPs:

• Stay informed and agile: This space is evolving monthly. Allocate time for your teams to experiment with AI tools and share learnings. What worked in Q1 2025 may need tweaking by Q4 2025.
  ‍
• Pilot your own AI assistants: Consider deploying chatbots on your site that use generative AI to assist shoppers. Not only can this improve conversion (and gather data on what people ask), it also prepares your organization for an AI-centric customer service future.
  ‍
• Focus on trust and transparency: In an AI-curated world, trust becomes currency. If your brand is the one consistently recommended by unbiased AI because of strong reviews and product quality, that’s invaluable. Double down on quality and customer satisfaction – happy customers feed the AI virtuous cycle (through good reviews and word-of-mouth).
  ‍
• Rethink metrics: Start tracking how much traffic or sales are coming via AI referrals (you can see referral domains like chat.openai.com or bing.com in analytics). It might be single digits now, but chart that trend. Set KPIs for AI channel growth and incorporate them into your reporting.
  ‍
• Educate and empower your team: Ensure your team (or agency partners) have the skills to create AI-ready content and analyze AI-driven insights. This might mean new training or hiring (as Gartner’s analyst said, “hire people who understand how AI affects search engines”). Investing in these skills now will pay off as AI’s role in marketing continues to expand.

The future of shopping is being written by algorithms and natural language, but the heart of it remains human: people seeking the best products and brands seeking to fulfill those needs. Generative AI is simply changing the interface and rules of engagement. As marketing leaders, if we remain customer-centric and data-driven, this revolution is not a threat but a tremendous opportunity. The brands that adapt will find that AI assistants can be among their best “salespeople,” delivering qualified customers practically to the checkout page. The agencies that adapt will solidify their role as indispensable guides through the new terrain.

In the end, those who innovate in this AI shopping era will capture new growth, while those who stick to old playbooks may see diminishing returns. The message is clear: adapt, experiment, and embrace AI’s potential – or risk getting left behind. With a sharp strategic focus and willingness to evolve, we can turn this disruption into a new engine of growth for digital commerce.