Here's the funny thing about how technology impacts marketing. Despite all the complexity we pile on top of it (the frameworks, the funnels, the attribution models), the job has always been the same: deliver the right message to the right person at the right time.

That's it. That's the whole game.

But here's what I learned at Sonos that changed how I think about this: most of the time, we're not the ones delivering the message. Our customers are. People who love a product tell their friends about it. Word of mouth happens in the dark, beyond our control, and that's actually fine. Our job is to make the message simple enough that it travels well. I didn’t market Sonos as a "modular redundant audio system for your home." Just "home sound system." Something a regular music loving person can actually repeat at a dinner party.

This is the insight that unlocks what's happening right now with AI.

The Semantic Transformation

A new participant has entered the word of mouth chain. And it's not human.

When someone asks ChatGPT or Google's AI Mode what digital picture frame they should buy for their parents, or which warehouse automation vendor fits their needs, the AI becomes the friend making the recommendation. It reads everything, synthesizes it, and delivers an answer. Not a list of links. An actual recommendation.

This is what I call Semantic Transformation. For twenty-plus years, we played the keyword game. We optimized for strings of text that triggered algorithms. The machines couldn't actually understand what we were saying (they just matched patterns). Now they can. AI systems understand semantics and sentiment. They can read your website, comprehend what you actually do, and then check that against what other people say about you on Reddit.

That last part is important. These systems have built-in BS detectors, calibrated not by personal experience but by sentiment analysis and authority signals. They're checking your claims against reality.

What AI Actually Wants

If you want to know what it feels like to be a robot crawling the web, just think back to the last time you tried to find a recipe online. You found a page that sounds like the exact dish you want to make but then you had to scroll through someone's entire life story about how their grandmother made this dish in 1973 and how it reminds them of summer vacations. And you're like: just show me the damn recipe!

That's how AI looks at most websites.

Humans love narrative. We crave stories. We'll take random events and weave them into meaning (hello, conspiracy theories). That's how we're wired, and it's why we create the content we create and constantly talk to our friends about it.

AI is different. It wants structure. Clear passages. Definitions. Entities and the relationships between them. Claims it can verify. When you bury your actual value proposition under layers of brand poetry, the AI struggles to extract what it needs. And when AI struggles, it goes somewhere else for the answer (or worse, makes something up).

Semantic Authenticity: How to Win

Here's where I get optimistic.

The keyword era rewarded gaming the system. You stuffed keywords, you chased algorithms, you played by arbitrary rules that had nothing to do with helping actual humans find the best solutions to their problems. It was a dumb game, and we all knew it.

The semantic era rewards something different: authenticity. Say what you do. Do what you say. Make claims you can back up. Organize your content around the problems you solve, not the features you ship. You still need to make it look beautiful and sound great to your core audience personas – which now also include robots.

I call this Semantic Authenticity. It means making your truth machine-readable. Building what my colleague Rhonda Lowry calls a "truth layer" that AI systems can confidently extract and cite. Creating content in the format people (and robots) are actually searching for, which is increasingly questions, not keywords.

The good news? This is actually just good marketing. It's problem-solution thinking. It's clarity. It's the stuff we always should have been doing but got distracted from by the SEO industrial complex.

Why This Year Matters

Don’t panic, marketing by humans is not going away. But, don’t sleep on this transformation.

AI systems are in the middle of forming their mental models of every category. They're learning what "digital picture frame" means, who the leaders are, what the trade-offs between options look like. These models are stabilizing. And whoever shows up clearly now becomes the default answer for years.

Think of it like semantic gravity. Right now, your category is still malleable. You can teach the model what this space means. Wait too long, and you're fighting entrenched assumptions.

This isn't about random experimentation or spending that new BoD-approved AI slush fund as quickly as possible. It's about recognizing a window. The companies that establish their semantic foundation now (clean up their truth layer, frame their category on their terms, create content that AI can reliably extract) will have durable advantages. Those that wait will spend years trying to displace someone else's positioning.

Where to Start

If you're a marketing leader wondering what to actually do, here's how I'd frame it:

First, recognize this isn't traditional SEO and it isn't a one-time optimization exercise. It's a transformation in how your organization thinks about expressing strategy in content and messaging. The semantic web requires semantic thinking.

Second, focus on foundations before tactics. Clean up your truth layer. Make sure what you say about yourself is clear, structured, and verifiable. Organize around problems and solutions, not just products and features.

Third, get specific about your category. How do you want AI to understand your competitive position? What comparisons matter? Start claiming that territory now.

For the detailed framework on how to execute this, I'd point you to my colleague Rhonda Lowry's work on PACE (Passage-level Clarity, Authoritative & Comparative Content, Coherent Semantic Ecosystems, and Entity Specificity). She's built the playbook for what machine-legible content actually looks like in practice.

But the strategic insight is simple: AI is now your friend's friend. Make sure it knows your story well enough to tell it right.

___________________________________

Michael Papish is Senior Vice President at IfThen, where he helps companies navigate the intersection of technology and marketing. Previously, he held leadership roles at Sonos and Markforged and co-founded MediaUnbound (acquired by TiVo in 2010) which applied machine intelligence to generating personalized music & video recommendations.