LLMs like ChatGPT are transforming how people discover your brand and products. Here are 3 ways to adapt

ChatGPT has become a go-to consumer resource for research, advice, writing, and more, according to a recent study from the National Bureau of Economic Research (NBER).

The study found that 70% of chats with the LLM’s 700 million weekly users are non-work-related, and that most people turn to ChatGPT for “practical guidance” or to “seek information”: to help them decide what to buy, for example, or how to solve a problem.

This shift is similar to the rise of search engines in the early 2000s, but this time, discovery is more conversational, more personalized, and more contextual. Consumers aren’t typing “best moisturizer for dry skin.” They’re asking, “I’m in my mid-20s and have combination skin. Can you recommend a skincare routine for me?”

ChatGPT and other LLMs have changed everything, from how writers write to how consumers search. In order to best position your business and your products, marketers need to adapt, too, to how LLMs are impacting consumer expectations.

Here’s how.

1. Make sure LLMs are aware of your brand

ChatGPT is now a consideration engine. The NBER research shows that about half (49%) of ChatGPT messages are classified as “asking” rather than “doing.” This means users are comparing options and seeking advice, many exclusively on ChatGPT, before jumping over to a brand’s website.

Unlike channels like email and SMS, you don’t have control over what users see in an LLM. But this is still a moment of influence for your brand. Whoever answers customer questions best in this research stage will be better positioned to win loyalty and trust before purchases.

It’s important, then, to make sure you’re prepping for ChatGPT queries by making your brand digestible for AI ecosystems. LLMs analyze massive amounts of information like customer reviews, detailed product descriptions, and category tags to deliver personalized recommendations that match user requests.

Here are a few ways to make sure your product catalog and supplementary content are AI-friendly:

• Generate structured content like consistently tagged product pages, exhaustive FAQs, and long-form pieces with headers that mirror real customer questions and provide clear, concise answers in the opening sentences. Structure each section as a mini-article that can serve as input for LLMs like ChatGPT.
• In your product descriptions and website content, answer questions (e.g., “Will this work with X?”), offer comparative context (e.g., “softer than cotton but more durable than polyester”), and include specific use cases and benefits. This contextual information helps LLMs understand exactly what to recommend and why.
• Use schema markup with accurate identifiers like Global Trade Item Numbers (GTINs) and Stock Keeping Units (SKUs), pricing, and availability data on your product detail pages. AI platforms rely on this kind of structured information to verify products are real, in stock, and available for purchase before recommending them.

AI marketing tip: Nothing hurts your AI shopping visibility faster than outdated inventory data. Connect to feed management services like GoDataFeed that can push product feed updates to multiple channels, including AI platforms. This way, when items are in stock, LLMs are more likely to recommend them.

2. Mimic the conversational feel of ChatGPT in your marketing

ChatGPT has trained consumers to expect instant, conversational answers to their questions. Remember, the NBER categorized nearly half of ChatGPT interactions as “asking”—seeking advice rather than just task completion.

Importantly, conversations in this category are consistently rated higher in quality than other categories, based on user satisfaction and feedback. This suggests people value AI as an advisor that helps them make decisions, more so than just a tool.

Your marketing content, then, needs to feel less like “I’m talking to you and selling you something” and more like “We’re having a helpful conversation.”

Here are a few ways to bring that ChatGPT-style conversational approach to your own marketing channels:

• Interactive product finders and quizzes: More than ever, brands need to find ways to create genuine connections with their customers. Create a quiz that asks a few questions and recommends the perfect match or helps consumers make more informed decisions.
• Flows that respond to actions: Use behavior to anticipate questions and proactively answer them by building flows that feel like a conversation unfolding organically over time. If someone views a high-ticket item multiple times, for example, trigger an email that addresses common objections (“Wondering if [product] is still worth the investment? Here’s what our customers say.”)
• Proactive help from AI shopping assistants: Implement an AI customer agent that can intervene when someone’s abandoning their cart or check-out, offering guidance on store policies, product recommendations, and other FAQs.

Ministry of Supply wanted to improve their customer experience with on-brand, dynamic account pages that complemented the shopping experience and offered access to live chat support. The officewear brand implemented a 24/7 AI shopping assistant that gives customers a choice between immediacy and a personal touch, with a “talk to a human” option throughout.

In just 60 days, the AI agent resolved 84% of product recommendation chat queries.

AI marketing tip: To support conversational marketing at scale, make sure you’re using channels that allow for two-way dialogue. RCS messaging and WhatsApp allow customers to reply to your messages and ask follow-up questions, just like ChatGPT.

3. Use AI to learn about customers and personalize accordingly

Just as customers expect ChatGPT to remember their preferences and adjust responses, they now expect the same from brands.

According to Klaviyo’s future of consumer marketing report, 74% of consumers expect more personalized experiences from brands. They want brands to know and meet them with relevant offers, at the right time, on their preferred channel.

ChatGPT-level personalization starts with a CRM that connects data from every touchpoint: email, text messaging, web chat, social, your customer hub, and more. This means bringing together purchase history, email clicks, text message replies, customer service interactions, and real-time conversations into a single source of truth with unified customer profiles that are always up to date.

Your CRM can collect and store customer data in multiple ways:

• Transactional data from purchases, returns, and subscription activity
• Behavioral data from email opens, SMS and WhatsApp replies, and push notification clicks
• Conversational data from AI agents that learn preferences through natural dialogue
• Zero-party data from customer hubs, surveys, and quizzes where customers explicitly tell you what they want

This unified data foundation is crucial for the kind of AI that can help you enhance your personalization efforts, including:

• Predictive analytics: Instead of treating all customers the same, use AI to anticipate individual needs, such as sending replenishment flows when customers are most likely to order again, or targeting customers who have high predicted customer lifetime value (LTV) with special offers.
• Channel optimization: AI can analyze each customer’s ideal sequence of email vs. SMS vs. push messages, then automatically route messages to their preferred channels.
• Timing optimization: AI can also identify when customers are most likely to engage and schedule messages for those optimal windows.

The Willow Tree Boutique started using predictive analytics to send campaigns to customers whose AI-predicted next purchase date was in the last or next 30 days, as well as campaigns highlighting pricey apparel to segments to customers with a predicted CLV over $500, or an average order value over $150. Now, 17.1% of their Klaviyo-attributed revenue comes from predictive analytics segments.

Dishware brand Corkcicle, meanwhile, segments marketing messages based on AI-powered channel affinity. They’ve discovered that their most efficient segment is customers who’ve engaged in the last 90 days who prefer email. “Channel affinity allows us to be efficient and targeted and has provided great results,” says ecommerce marketing specialist Erica Olsen.

AI marketing tip: Not sure where to start with personalized marketing? An AI marketing agent can help by auto-generating on-brand, personalized experiences for email, SMS, WhatsApp, and more, powered by your customer data.

Turn curiosity into conversion with K:AI (Klaviyo AI)

NBER’s research makes it clear that ChatGPT is the new consumer operating system. For marketers, it’s an opportunity to make sure your content, data, and customer experiences are not only ready for AI to find and understand, but also tailored for new consumer expectations.

K:AI (Klaviyo AI) helps you meet those expectations and deliver the same personalized, conversational experience in your marketing channels.

• K:AI Marketing Agent creates a marketing plan and ready-to-send campaigns, flows, and forms that help you stay connected to consumers once they discover your brand.
• K:AI Customer Agent keeps that same intelligence and empathy alive after the click—solving, recommending, and retaining in real time.
• The omnichannel campaign builder empowers you to build multi-step campaigns across channels that drive action with more interactive messages, powered by AI that learns how customers engage.

As discovery becomes conversational and more and more purchases begin with a prompt, Klaviyo makes sure your brand shows up with purpose—turning “Who should I buy from?” into “I’m glad I chose you.”