Mark Kabana, VP Of Data Innovation, Yext On Local SEO, AI-Driven Search, And The Future Of Data Innovation

Mark Kabana: Over the past decade, local search has definitely evolved, and while it may have started with listings, we’ve known for a long time that wasn’t the whole game. Back in 2016, Places Scout published the first local ranking studies based on actual data that showed reviews and photos were just as critical to rankings as basic NAP consistency.

So, while the broader industry may have treated local SEO like a “listings-only” problem for years, we’ve always viewed it as an interconnected ecosystem. What’s happening now with AI is really a validation of that early view, and it has raised the bar across the board.

AI exposes inconsistencies instantly, and it rewards businesses with clean, complete, and well-structured consistent data across every digital surface. Not just directories, but also your website, social channels, review platforms, and anywhere your business is mentioned.

It’s no longer enough to just be present online. It’s about being understood and trusted by both people and machines. That means schema markup, review content, brand mentions, and a smart data strategy all working together.

The brands winning today are the ones that treat their digital presence like a living system that is constantly monitored, improved, and optimized at scale.

Mark Kabana: We’re entering a new phase where the focus is shifting from insight to execution. Traditional automation tools could tell you what happened, like which keywords dropped, where you were outranked, or how many reviews you received. The next frontier is about understanding why it’s happening and what to do about it, in real time and at scale.

AI is pushing the industry from passive reporting to active recommendations. Soon, it'll be fully agentic with autonomous systems that can both flag underperformance in a specific region and take immediate action by updating listings, optimizing landing page content, and even responding to reviews using brand-safe language.

We built Places Scout to surface hyper-local patterns. Now, connected with Yext’s broader AI and execution layer, we’re starting to see that intelligence turn into immediate, localized action. That’s the frontier.

Mark Kabana: One of the most underrated aspects in local SEO is brand prominence, and I don’t just mean backlinks. I’m talking about unlinked brand mentions, repeated name references on trusted sites, local PR coverage, and even social chatter.

Google and large language models already treat businesses as entities, not just websites. They rely on signals that reflect how well-known and trusted those entities are within a specific category or geography. That could be a mention in a neighborhood newspaper, a Reddit thread, or a partner listing on another company's site.

Another overlooked factor is imagery. Too many businesses still rely on stock photos, but both users and AI can now easily spot the difference. Real, high-quality, and relevant visuals help build trust and improve visibility.

These things used to be considered secondary. Now, they are essential signals that directly influence rankings and visibility across both traditional search and AI-driven experiences.

Mark Kabana: AI is incredibly powerful, but it is only as good as the data it is trained on and the guardrails you put around it. When companies move too quickly into full automation, they risk diluting their brand voice or unintentionally reinforcing bias.

We are already seeing tools that write responses, edit content, and even recommend products. But if those decisions are based on flawed or incomplete data, the result is scaled misinformation or inconsistency.

The key is pairing AI with strong governance. Use automation to handle repetitive tasks, but make sure there is human oversight where nuance, empathy, or judgment is required. Most importantly, be transparent about how data is collected, processed, and used. That transparency builds trust, and trust is what keeps customers coming back.

Mark Kabana: We are entering an era where rankings do not necessarily equal visibility. You could technically rank for a keyword but never appear in an AI Overview. Or you might not show up in the top three search results but still be cited in an AI-generated answer, which often appears above everything else.

Backlinks and rankings still matter, but they are no longer the full picture. What matters now is being cited and trusted, showing up in the structured data that powers AI answers, and having your content appear in the sources large language models pull from.

We are starting to see new metrics emerge, including mention frequency, entity recognition, structured data coverage across directories, and engagement signals across platforms. This is a shift from rank and click to appear and influence, and it is reshaping how we define success in search and AI responses.

Mark Kabana: The power of personalization is undeniable, especially when it is local. But the line between helpful and intrusive often comes down to consent and context.

As AI systems start to retain memory and personalize results based on previous interactions, we are seeing what we call ranking fragmentation. This is when results vary dramatically based on what the AI already knows about you.

For example, if your child has a shellfish allergy and you ask for a restaurant recommendation, the AI might first ask if your child is joining you. If the answer is yes, it will exclude seafood restaurants entirely. That kind of personalization can be incredibly helpful, but only if it is done with care.

Brands should use location and intent signals to improve relevance, not to manipulate behavior. That means surfacing the right offer at the right time, without making assumptions that feel invasive or overly personal.

Transparency is essential. Be clear about what data you are collecting, ask for explicit consent, and always provide real value in return. Hyper-personalization can build trust, but only when it feels like a service, not surveillance.