For practice owners with great reviews and a quiet phone
Patients have started asking AI — ChatGPT, Google's AI Overviews, Perplexity — "who's the best chiropractor near me for sciatica?" And AI gives them one answer, not ten blue links. If that answer isn't you, your reviews, your website, and your Maps ranking never even get seen.
The AI Visibility Snapshot shows you exactly what the machines say about your practice, why they say it, and what it's costing you in new patients per month.
Try it yourself, right now
Did it name you? Did it name a competitor? Did it recommend a physical therapist instead?
That answer — not your Google ranking — is what a growing share of your next patients sees first. In 2026, fewer than a third of Google searches still send a click to an outside website; the rest are answered on the results page itself (SparkToro, 2026). The map pack you've been fighting over is shrinking real estate.
What changed
For fifteen years the formula was stable: get reviews, rank in the map pack, run some ads, and the phone rings. You executed it. Most docs reading this have the reviews to prove it.
Then two things happened:
Here's the uncomfortable part: the signals AI engines use to decide who to recommend overlap with — but are not the same as — the signals that win the map pack. Structured data, entity consistency, how your services are described in plain patient language ("back cracking," "adjustment") versus clinical language, whether your hours, booking, and conditions treated are machine-readable, and what third-party sources say about you.
A practice can be #1 in the map pack and invisible to AI. Most are. Nobody's checking, because almost nobody in local healthcare marketing knows how to check yet.
Before you pattern-match this
If you've owned a practice for more than three years, the odds are you've been through some version of this cycle:
An agency promises new patients. Month one is great — 20, 25 through the door. Month two: 15. Month three: you're chasing your rep for a call. By month six you're paying a retainer for what amounts to web hosting.
That decay pattern isn't bad luck and it usually isn't fraud. It's structural: most agencies front-load results by burning your offer through a cold audience with discounting, then have no second act because they never fixed the underlying asset — your actual visibility and conversion infrastructure. When the ad novelty fades, so do the numbers. The agency model needs your retainer more than your retention.
This is why the Snapshot is not a retainer. It's a diagnostic with a fixed price and a deliverable you own. You can take the report and:
There is no version of this where you're locked into anything. The diagnosis is the product.
The AI Visibility Snapshot
The Snapshot is a fixed-scope diagnostic of how AI engines and modern search see your practice — and where prospective patients are leaking out between "searched" and "showed up."
Direct queries against ChatGPT, Google AI Overviews, Perplexity, and Copilot for the searches patients in your area actually run — by condition ("sciatica," "neck pain after car accident"), by intent ("best chiropractor near me," "chiropractor open Saturday"), and by brand ("Dr. [Name] reviews"). You see the actual outputs, screenshotted and dated. Who gets recommended, who gets ignored, and what the engines get wrong about you.
The signal audit: your structured data, entity consistency across the web, how your services are described (clinical language vs. the words patients actually type), citation sources AI engines trust, review signal readability, and the machine-readability of your hours, booking, and conditions treated.
Visibility means nothing if the visit doesn't convert. We trace the patient path on a phone: how many taps to book, whether booking is real-time or a "we'll call you back" form, what happens to an after-hours call, and how fast a web inquiry gets a response.One practice owner put it perfectly: when the waiting room is fullest, the most calls go missed.
Findings translated into the only math that matters: estimated new patients per month at stake, what each leak costs at your average case value, and a prioritized fix list ranked by patients-recovered per hour of effort — not by what's easiest to bill for.
The numbers
You know your math better than any marketer, so here it is in your units — deliberately conservative. If your average new patient is worth, say, $400–$800 in collections over their care episode (cash practices and high-PVA offices will be well above this), then:
The Snapshot doesn't ask you to take these numbers on faith. It measures yours and shows the working.
A lot of what gets pitched to practice owners — discounted new-patient specials, review incentives, prepaid plan promotions — sits in legal gray zones around patient enticement, dual fee schedules, and review-platform rules that vary by state. You've probably had a vendor pressure you toward something your gut flagged.
Every recommendation in the Snapshot is screened against that reality. Where a tactic has a compliance edge in Georgia, the report says so explicitly and gives you the safe version. Nothing in this engagement will put you in a conversation with your board.
About
I'm Matt MacLeod. For 17 years I've done search strategy at enterprise scale — currently leading organic and AI search across a 50+ property international hospitality and entertainment portfolio, previously running a 77-country site migration for a global software company. My day job is figuring out how Google's and the AI engines' machinery actually decides who gets recommended, at a scale where being wrong costs real money.
Keystone Strategy AI is me bringing that machinery down to local practices in North Georgia and the Atlanta metro — starting with the niche where the gap between "great practice" and "visible practice" is widest.
Two things I am not: I'm not a practice management company, and I'm not going to teach you scripts. I diagnose search and conversion infrastructure. You run your practice.
I'm based in Dawsonville, a member of the Dawson County Chamber, and I do these in person where geography allows. If we're within an hour of each other, the walkthrough call can be a walkthrough visit.
Fixed scope, written report, prioritized fix list, 30-minute walkthrough. Delivered in ~5 business days. No retainer attached, no contract to exit, no upsell required to get the full findings.
Run my AI Visibility Snapshot →If the report shows work worth doing and you want me to do it, implementation is scoped per-project or month-to-month — and the Snapshot fee credits toward it. If you'd rather hand the report to your web vendor, your nephew, or another agency: genuinely fine. It's built to be executable by whoever you trust.
The questions you'd ask on Reddit before buying
The Snapshot itself generates zero — it's a diagnostic, and anyone who promises you a patient count from an audit is doing the thing you've learned to distrust. What it does is quantify, in your numbers, where patients are currently being lost and what each fix is worth. Whether those fixes happen — by you, your vendor, or me — determines the patient count. The report shows the math for every estimate it makes.
There's no contract. The Snapshot is a one-time fixed fee. If you later hire me for implementation, that's month-to-month or per-project, cancelable anytime, and you keep everything built. I don't host your website hostage and I don't own your accounts — everything lives in accounts you control.
The scope is the four panels listed above, and the deliverable is a written report with dated screenshots. You can compare the two yourself. There's nothing to take on faith and no "we're working on it in the background."
Some aren't — yet. But their adult kids researching care for them are, and AI Overviews are already injected into the regular Google results your patients do use, whether they know it or not. The Snapshot also covers the unglamorous leaks — booking friction, missed calls, response time — that cost you patients today regardless of what you believe about AI timelines. If the AI panels come back showing your market hasn't shifted yet, the report says that too, and you'll know before your competitors do.
You can run the queries yourself — the top of this page tells you to. What you can't easily do is the signal audit underneath: knowing which of the dozens of possible factors actually moved the answer, which fixes are load-bearing versus cosmetic, and what's changed in how these engines source local health recommendations in the last six months. That's the part that's a full-time job. Mine, specifically.
Because the gap is widest here. Practices in this profession over-invested in reviews and websites relative to almost any local category — and the search layer shifted out from under that investment harder than anywhere else. Also: it's a profession full of owners who did everything right and got punished for it, and fixing that specific injustice is satisfying work.
One fixed fee. One written report. Five business days. Yours to keep, whoever implements it.
Run my AI Visibility Snapshot →Prefer to sanity-check me first? Run the ChatGPT query from the top of this page, then email me the screenshot — I'll tell you the single most likely reason for what you're seeing, free, no call required.
