The Agent Freeway: What Happens When Patients Stop Searching

It's 2028. Sarah's molar cracks at 9pm.

She doesn't open Google. She doesn't ask her group chat. She doesn't read a single review.

She tells her phone: "find me a dentist."

Her agent queries 47 practices in her network within 200 milliseconds. It filters by her Delta Dental coverage, cross-references her calendar, weights the results against her review preferences, and books her in for 7am Tuesday.

She never sees a website. She never compares three options. She never even sees a search result page.

The agent did all of it.

Your website had an audience of one — and it wasn't a human.

This is the future I want to talk about. Not in a sci-fi way. In a "this is happening faster than you think and your marketing is about to break" way.

I call it the Agent Freeway. And if you run a dental practice, you have about 18 months to figure out how to get on it.

This Isn't Sci-Fi. It's Already Half-Built.

Every piece of this future already exists. They just haven't fully connected yet.

• AI Overviews are already eating organic clicks. Pew Research found that when an AI summary appears, users click a link only 8% of the time. Without the summary, they click 15% of the time. That gap is widening every quarter.
• ChatGPT shopping, Perplexity Pro, and Claude's web search already answer most "should I buy this" questions without ever showing the user a website.
• Anthropic's MCP (the protocol that lets AI talk to tools) and Google's A2A (agent-to-agent communication) are shipping right now. These are the road and the lane markers of the Agent Freeway.
• Stripe and Skyfire are building payment rails so agents can transact on your behalf without a human pulling out a credit card.

Every layer is being built. The question isn't if this connects. The question is who's ready when it does.

I wrote a few weeks back about agentic AI coming for dental marketing. That post was about AI agents running your marketing for you. This post is about something bigger and weirder: AI agents replacing the patient as the audience for your marketing entirely.

The Internet Is About to Split in Two

Here's the framing that clicked for me.

The internet today does two jobs at once. It entertains us, and it helps us get things done. Watching Netflix and booking a dentist happen on the same internet, through the same browser, against the same Google search box.

That's about to end.

The internet is splitting into two planes:

The Human Plane. Entertainment, social media, video, browsing for fun. The stuff humans actually want to do with their attention. This layer keeps growing.

The Agent Plane (the Agent Freeway). Transaction, execution, problem-solving. The boring "get things done" layer. This goes underground, agent-to-agent, machine-speed.

Humans stop using the internet to find a dentist, book a flight, or compare insurance plans. They use it to watch a video and laugh at memes. The "getting things done" layer becomes a back-channel highway that humans never see.

And on that highway, your website isn't a destination. It's a data source. Or it isn't there at all.

Why This Breaks Traditional Dental Marketing

Let me get specific about what stops working.

SEO. Ranking #1 means nothing when no human ever sees the search results page. Agents don't care about your meta description or your H1. They want structured, queryable data and they want it in 200ms.

Google Ads. You're paying for impressions when the impression is a bot that filtered you out before any human saw the ad. The entire pay-per-click model assumes a human eyeball. Take that assumption away and you're lighting money on fire.

Pretty websites. Your hero video, your stock photo of a smiling family, your "about us" page with the inspirational quote. None of it gets parsed. Agents skip past the design and pull the data underneath.

Reviews. Still matter. But agents weight them differently than humans do. A human reads three reviews and makes a gut call. An agent reads all 847 of them, scores them semantically, weights for recency, and outputs a confidence score. The "5 stars on Google" shorthand stops being shorthand for anything.

Referrals. The agent is the referrer now. Your patients aren't telling their friends about you. Their agents are telling their friends' agents about you, based on data signals you may or may not be sending.

This isn't doom. It's a structural shift. And the practices that see it coming are about to have an enormous advantage over the ones that don't.

What Wins on the Agent Freeway

So what does win? Let me walk through the new moat.

Machine-Readable Everything

Schema.org structured data was the appetizer. The main course is real, queryable agent endpoints. Your hours, your services, your prices, your insurance acceptance, your real-time availability — all of it needs to be machine-legible at sub-second speed.

The practice that responds to an agent query in 200ms with clean structured data wins. The practice with a beautiful website and no API loses, even if the website is gorgeous.

Verified Signals

Agents don't trust marketing copy. They trust verified data. NAP (name/address/phone) consistency across the web becomes a trust signal. Verified insurance partnerships become a ranking signal. Real-time availability becomes a ranking signal.

I wrote about why citations and NAP consistency still matter for AI search. On the Agent Freeway, those same signals become load-bearing in a much bigger way.

Reputation That Agents Can Parse

Star ratings get commoditized fast. What matters next is the content of reviews. Agents can read what's actually being said: did people mention the dentist's bedside manner? The wait time? The way the front desk handled insurance billing? Those become rankable variables.

The Real Unlock: Affinity Matching at Machine Speed

This is the part most dentists are going to miss, and it's the most important one in this whole post.

Today, if a patient wants a dentist who's an avid Lakers fan with a family, what do they do?

They ask their group chat. They stalk Instagram. They read 60 reviews hoping someone mentions vibe. Most people don't even try. They just pick whoever's closest and in-network and live with the awkward small talk.

The agent does it in 200ms.

It cross-references:

• The doctor's bio page ("dad of three, season ticket holder since '08")
• Practice Instagram (kids' Halloween costumes, courtside selfies)
• Podcast interviews and local press
• Review semantics ("Dr. Kim talked about his daughter's soccer game the whole appointment, I loved it")
• Community signals (sponsors of the local little league, mentioned in a Lakers fan subreddit)

A patient says: "find me a dentist near me, in-network, who has kids and is a Lakers fan." The agent returns three matches, ranked. Done.

This flips conventional wisdom on its head. Most dentists are going to optimize for clinical signals when this future arrives. Services. Hours. Insurance. Prices. That's what feels professional. But agents already have that data. It's table-stakes. It doesn't differentiate anyone.

The differentiation moves up the stack. To personality, affinity, and shared identity.

The sterile corporate website with the stock photo of a smile? Loses. The practice that posts the doctor at his kid's recital, the Lakers game, the team Halloween costume? Wins. Not because humans see those posts. Because agents can parse them and match patients to them.

Personal brand for dentists doesn't matter less in the agent era. It matters dramatically more. It becomes the only thing left to differentiate on after the table-stakes data gets commoditized.

The practices that go all-in on the doctor as a real human being — hobbies, family, opinions, community involvement, podcast appearances, the whole thing — are building the exact signals agents will use to match patients. The practices hiding behind a generic "trusted family dentistry" website are building nothing.

What Dentists Should Do This Year

I'm not going to tell you to panic. The human plane still exists, and it's going to keep existing for years. Patients will keep finding you the old way for a while. But the practices that wire up for both planes now will own the next decade.

Here's what I'd do if I were running a practice in 2026:

1. Audit your structured data. Most dental websites have garbage schema markup. Get it clean. This is the foundation everything else sits on. I cover the technical side in my insurance pages and dentist-near-me posts.

2. Clean up your NAP across the agent-readable web. Citations still matter. They just matter for a new reason now. Agents use them as trust signals.

3. Make the doctor a real person on the internet. This is the big one. Bios with hobbies and family. Social media with personality, not just before-and-afters. Podcast appearances. Community involvement that gets covered in local press. Agents will read all of it. Patients (via their agents) will match on it.

4. Don't fire your marketing team. The human plane is still your trust-building layer. Brand still matters. Reputation still matters. Patients are going to keep judging you the moment they walk in the door, no matter how they got there.

5. Pick a partner who's building for both planes. Most dental marketing agencies are still optimizing for a 2018 internet. The ones thinking about agentic search, machine-readable data, and personality-driven brand are the ones to bet on. I'm obviously biased here — that's what we're building toward at Lasso MD — but even if you don't work with us, find someone who at least thinks this way.

The Window Is Open Right Now

The practices that wire up for agents in 2026 and 2027 own the next decade.

The ones who wait until 2029 are competing on a freeway they didn't build a ramp for.

This is the same bet practices made about websites in 2003 versus 2013. The early movers got cheap, dominant SEO that took competitors a decade to dislodge. The late movers paid 5x for the same outcome and never caught up.

The Agent Freeway is going to be the same story, except the timeline is compressed. The early movers have 18 months, not 10 years.

Your patients aren't going to stop using the internet. They're going to stop searching it. Their agents will do that for them. The question is whether your practice is on the freeway when they pull up — or stuck on a frontage road nobody drives anymore.

If you want to talk about what this looks like for your specific practice, book a call. I'd rather have this conversation now, while there's runway, than in 2028 when the lanes are already full.