Be the Answer: Yext Leaders on AI Visibility, Trust, and the Future of Marketing

TL;DR: In a world where AI search rewards consistency and structured data across every digital surface, fragmented information gets cited wrong (or not at all). 'Being the answer' in AI search means becoming the data the model reaches for when it needs to be confident. That's a more durable position than any ranking.

---

AI search has fundamentally changed how customers ask questions, find information, and choose brands. But when it comes to driving AI visibility, the how is a lot more complicated than the what. Read five articles about what it actually takes to help your brand show up in search today — to earn AI citations, stay trusted, and win the customer journey — and you might get five different takes on the mechanics.

To dive deeper, we sat down with five Yext executives — Wendi Sturgis, Christian Ward, Chris Brownlee, Rebecca Colwell, and Kamelia Gouchev — to hear where they think the industry is, what brands still aren't doing, and what it really means to "be the answer" when the answer layer is increasingly an AI engine. Their different roles influence their answers in fascinating ways — from Wendi's "big picture" focus as our global sales lead to Christian's technology deep-dive as our chief data officer.

Their responses are below, lightly edited and condensed for clarity.

Lauryn Chamberlain, Senior Editorial Strategy Manager: What shifts in AI search are brands still not taking seriously enough?

Wendi Sturgis, EVP, Global Sales: I'm actually seeing brands start to get curious and recognize this is a serious shift. The issue is that the space is so new, people are taking a cautious, test-and-learn approach while they watch a few things: what others are doing, what their trusted partners and agencies are telling them to do, and how their current work impacts their discoverability. It's a wait-and-see mode.

Rebecca Colwell, SVP, Marketing: The biggest mistake brands are making right now is jumping straight into execution, chasing shortcuts for AI visibility, without asking the harder question: are we actually structured to show up consistently? AI search rewards accuracy and consistency across every digital surface. That can't happen if paid and organic teams are siloed, or if corporate and local are working from different sources of truth.

Brands keep treating AI visibility like a new channel to add to the mix, but it's really an operating model problem.

Chris Brownlee, SVP, Product: For the last 15 years, many brands have had different teams managing the many channels their digital presence shows up in. A team for review responses. A team for social media. Another for websites and listings. In the AI search era, your brand's presence is an aggregate of all that information. AI search relies on web crawlers to gather information about your brand, and it gives answers based on what it finds across the entire web: your website, your listings, reviews left about you on sites you may not even be aware of, and even your competitors' data.

The strategy for the AI search era is understanding how you appear in AI search, where it's collecting information from, and taking a proactive approach to managing and growing your presence.

The other aspect rarely talked about: in traditional search, all you really had to make a match with a customer search was a keyword. But AI search knows a lot about the customer who's searching. It knows their demographics. It knows their hobbies. It has an immense amount of detail on the customer from all the previous conversations they've had with the AI search engine. To win in this new era, a brand needs just as much depth and breadth of data about itself to have the best chance of a strong, high-intent match with that customer.

Christian Ward, Chief Data Officer: What most brands still underestimate is how much the query itself has changed. A decade ago, a search engine inferred intent from signals like location, device, and rough demographics. Today, AI experiences can carry much richer context about the person asking: preferences, past interactions, location, loyalty status, dietary needs, household details, or other remembered signals. The consumer is no longer just sending a keyword. They are sending a keyword plus context. Brands still designing around keywords and impression counts are a cycle behind the consumer.

The other shift is real-time eligibility. Ad auctions were built around scheduled inventory and bid tables. AI wants a correct answer in the moment. If your pricing, availability, locations, hours, and service windows are not accessible when the model needs them, you can fall out of consideration even when the customer is looking for exactly what you offer.

Kamelia Gouchev, SVP, Client Success: The biggest miss is that brands are still optimizing for visibility, not inclusion.

Traditional search was about ranking on a page. AI search is about whether your brand is selected, interpreted, and cited in a generated answer. There is no page two anymore. There is presence or absence.

But the deeper shift is that we have moved from keywords to corroboration. AI models aren't just reading the brand website; they are cross-checking the brand across dozens of sources and looking for a consistent version of the truth.

I used to search for the best brunch near me and scroll through options. Now I ask, "What is a great brunch spot nearby for a casual Sunday?" and get two or three recommendations. If you are not in that answer, you are not in the running.

When a customer asks an AI a question and your brand isn't in the answer, what did you lose — and do you even know it happened?

Wendi Sturgis: That's part of the issue. We lack metrics, search volume, and data of really any kind. Marketers know the broader usage percentages, but they don't have clear metrics to tell them what's lost. This is an emerging space, and the best marketers understand that. They're willing to fight for investments despite the lack of full clarity. Many others are not.

Rebecca Colwell: Put simply: you lose a potential customer. You lose being included in a prospect's entire purchase journey.

Chris Brownlee: If the customer journey starts in AI and you don't appear, the race was lost at step one.

Let me give this some context. Everyone knows a website is valuable for customers to learn more about a brand and their services, offerings, and products. If that website isn't indexed by the AI crawler, it may as well not exist. That's why at Yext, we've put so much focus on schema markup and experimenting with other techniques like llms.txt. You're only visible to customers if you're visible to AI first.

Now, did you even know it happened? The AI search engine won't tell you. There are no analytics to check. What if you could compare your web logs to your competitors'? What if you could see who the main referrers are to your site and to your competitors' sites? If I knew my competitor's website was getting 4x the traffic from AI search engines, either from crawlers or real traffic, I'd know I'm losing.

Christian Ward: In classical search, I at least lost an impression, and I could often see the drop-off in logs. In a generative answer, there may be no impression, no referrer, and no clean trail to inspect. I lose the customer and the feedback loop at the same time.

This effect also compounds. If my brand is missing from the answer, the customer may never evaluate it, click it, review it, or mention it. That means fewer downstream signals for the brand and fewer chances to become part of the consumer's remembered consideration set. If my brand never makes it into that memory layer, the next prompt can filter me out before the question is even asked.

The practical answer is to build an evaluation set of the prompts that matter to your category and check it on a cadence. Without that, you are guessing about the surface that is shaping demand.

Kamelia Gouchev: You did not just lose a click. You lost the moment of consideration.

Take something simple. Someone asks "What is the best sushi spot near me for a date night?" The AI gives three options. Most people choose from that list and maybe double check reviews. They are not opening ten tabs anymore.

If you are not listed, that customer likely never discovers you.

The same is true in retail. If someone asks, "Where should I buy an affordable but good quality couch?" and a few brands are named, you are out before the decision even starts if you are not included. And in an AI-driven, zero-click environment, you do not even know it happened. You are being filtered out before the customer ever interacts with you.

There is no metric for being excluded from the answer. You only see the outcome, which might look like a gradual drop in traffic or fewer new customers with no clear explanation.

In a world where an AI model is a frequent touchpoint between a brand and a potential customer throughout a journey, what does it take to actually earn trust?

Wendi Sturgis: You need to control your brand everywhere consumers search. We've always emphasized that, and now more than ever. We're hearing that LinkedIn may matter, which means brands with distributed models need to make sure the channel is in control. TikTok has its own AI platform now. The channels where consumers interact keep proliferating, and you have to have a strategy to manage them all.

This isn't the top-of-funnel, sexy brand marketing. It's table stakes. As consumers move down the funnel, the decision-making process requires more detailed information, and you need to control that data everywhere.

Rebecca Colwell: Trust with an AI model works the same way trust with a customer does: it's built on consistency. If your brand information is accurate and the same everywhere it appears, models learn to rely on it. If it's fragmented or outdated across sources, you get cited wrong — or not at all. Earning trust in AI search is a data quality problem, not a content problem.

Chris Brownlee: Earning trust with AI search engines is all about having a consistent presence on the web. One way LLMs reduce their chance of hallucinating a response is by looking for signals that verify the data they see. If a brand has hours posted on Google that don't match MapQuest or Yahoo or the website, the model won't know which data point to trust. It will likely disregard it wholesale if things are too far askew.

Similarly, customers are still hesitant to fully trust these AI search engines. We've all experienced a confident hallucination from an LLM. It's easy to spot those hallucinations on a topic you have expertise in, and it makes you wary of the information you're getting back on topics you know less about, like store hours. Just as consistency is key for the AI search engine, the same is true for customers.

Christian Ward: Trust in this environment has two audiences. The model and the human. They are connected, but they do not evaluate trust in the same way. I think of that as the Trust Paradox.

Models look for reliability across the public record. Is the brand information consistent. Is it current. Is it supported by credible sources. Does the same version of the business show up across listings, reviews, local pages, websites, and third-party references.

Humans do something similar. They ask the AI for an answer, then check the brand's site, reviews, Google profile, photos, pricing, and availability before they act. The AI answer may start the trust process, but the surrounding ecosystem has to confirm it.

That means trust is not only a content problem. It is a data quality problem, a governance problem, and a customer experience problem. If the model says one thing, the website says another, and the store experience says something else, trust breaks on both sides.

The brands that win here will manage one reliable version of the truth across every surface. The answer the model gives about you should match what the customer sees when they verify it, and what they experience when they walk in.

Kamelia Gouchev: Trust now operates on two levels. First, earning trust from the model — and second, earning trust from the customer who is validating the answer.

To show up in AI responses, brands need consistent, structured data across all endpoints; high quality, factual, and unambiguous content; and strong external signals such as reviews and third party validation.

But more importantly, you need to signal that your information is the authoritative version. AI models are constantly comparing sources. If there is inconsistency, they do not know what to trust, and your brand becomes a risk to include.

But that is only half of the equation. Because users do not stop at the AI answer. They verify. When they click through, your website, listings, reviews, and local pages… they all need to reinforce the same story.

For example, someone asks, "Is this gym good for beginners?" The AI says yes and includes your brand. But when they check your site, it is filled with intense fitness imagery, and reviews mention that it feels intimidating. That disconnect breaks trust immediately.

Or a restaurant is described as family friendly, but there is no kids menu and reviews say otherwise. That gap is enough to lose the customer.

If there is a mismatch, trust breaks instantly. And once it breaks, it is very hard to recover.

Our recent study found that 73.8% of AI users trust AI recommendations, but nearly all of them still verify information elsewhere before acting. What does that mean for how brands need to show up?

Rebecca Colwell: I pay more attention to what people do than what they say. The fact that most AI users still verify before acting tells me that trust isn't fully formed yet. Some of that is habit; people have searched on Google for two decades, and that pattern doesn't break overnight.

But the practical takeaway for brands is this: AI is where customers discover and validate options. Google and your own properties are where they confirm. You need to show up well in both environments, and they need to tell exactly the same story.

Chris Brownlee: This comes back to customers still building trust in the LLMs behind the AI search engine. While the technology improves every day, the stories about LLMs recommending bad information keep coming. From the absurd, like "eating at least one small rock per day for minerals," to the dangerous, like Air Canada's chatbot giving incorrect information that led to them being sued. LLMs don't "know" facts. They predict the most likely next word. If the training data is sparse, they fill the gaps with plausible-sounding but completely fabricated info.

We're still in the relatively nascent days of AI search. The transition from traditional web to AI is still in progress. Trust signals are just as important for consumers as they are for the AI search engines.

Christian Ward: I read the verification behavior as the Trust Paradox showing up in the real world. A consumer gets a recommendation from the machine, then checks Google, reviews, the brand site, photos, pricing, and availability before acting. They are not blindly trusting the answer. They are auditing it.

That matters because the verification step is fast. The consumer is comparing the AI answer against every other surface they can see. If the AI says the restaurant is family friendly, but the reviews say it is loud and the site has no kids menu, trust breaks.

If the AI says a gym is good for beginners, but the photos and reviews suggest otherwise, trust breaks. Memory adds another layer. Once an AI recommends a brand, that recommendation can shape what the consumer remembers about the category. If the verification step confirms the answer, the brand gets stronger in the consumer's mind. If it contradicts the answer, the brand loses credibility.

For brands, the work is to reduce contradiction. The same facts need to hold across the answer, the listing, the local page, the reviews, the site, and the actual experience. AI may open the door, but verification decides whether the customer walks through it.

The more personal AI becomes, the less room there is for brands with weak or inconsistent signals. If the system already knows the consumer's preferences, context, location, and past behavior, the brand has to be clear enough to fit that context. Data consistency becomes a trust issue, not just a visibility issue.

Kamelia Gouchev: It means AI is the front door, but not the final decision maker. So brands need to win in two places: in the answer, to be considered… and in the broader ecosystem, to be chosen.

You can see this in how people actually behave. Someone asks, "What is the best nail salon near me?" The AI suggests a couple of options. The next step is to check reviews, look at photos, and maybe glance at pricing or booking.

If anything feels off, such as low ratings, outdated information, or a confusing website, they move on — even if you were recommended.

This is where the transition matters. Brands should not just optimize for the AI answer. They need to optimize for what happens immediately after it.

Finally, what does it mean to you to "be the answer"?

Wendi Sturgis: It means having a clear content strategy that lets you show up in the right ways, with the relevant content about your brand, your products, your services, and all the associated brand assets. Everything.

Chris Brownlee: To "be the answer," I interpret this as being the supplier of the data the LLM was trained on. There's a general sense that having your brand's information serve as training data is bad. For a marketer, it's the opposite. If my brand's blog post becomes the main source of truth when the AI search engine needs to predict the next best word, and the citation it serves leads to my website, then not only am I the answer, I become the authoritative truth.

Christian Ward: For me, being the answer means becoming the source the model reaches for when it needs confidence.

Written content still matters, but the more durable asset is the structured record a brand maintains across the open web. The entity graph. The canonical facts. The local pages. The inventory, pricing, hours, services, and availability that tell the model what is true right now.

If the model can reach your data, understand it, and verify it against other credible signals, you are no longer just competing for a ranking. You are supplying the answer.

Location is still a major part of this. Many high intent questions are local. The cleaner and more consistent entity often wins because the model has less ambiguity to resolve.

The longer term goal is to become part of the consumer's remembered consideration set. When they ask again next week, the brand should already fit the category, the need, and the context. That is a stronger position than a temporary ranking.

Kamelia Gouchev: It means your brand is the most reliable, consistent, and clearly understood source for a given intent.

There is a difference between being a result and being the answer. A result says, "We might have what you need." An answer says, "This is the right choice for your specific situation, and here is why."

Not just visible. Not just present. But the one the AI is confident enough to recommend.

Rebecca Colwell: Being the answer means showing up every time a potential customer has a question your brand can actually help with. Branded or unbranded. Early research or late consideration. On every channel where they're looking. The goal is simple: no relevant question goes unanswered by your brand, wherever it gets asked.

Ready to act? Find out if AI engines — and customers — trust your brand. Get your visibility report now.