How do patients find a med spa in Charlotte?
The short answer
Somebody in Charlotte decides they want Botox, or a consult about the laser a friend had, and the next thing they do is pick up their phone. From there the search runs down one of three routes. The map Google puts at the top of a local search. The ordinary list of results under it. Or an answer from an AI assistant that names a few practices out loud. That is how patients find a med spa in Charlotte, and in Greenville or Charleston, because the machinery is the same everywhere and only the names in the results change.
This piece is a walkthrough of that machinery, not a measurement of it. There are no figures here, on purpose. We do not have Charlotte search volumes or click-through rates and we are not going to invent them. Anyone who quotes a number for how patients split across those routes is either Google or guessing. What we can describe is the mechanism: what each route reads, what it decides on, and which part of the decision the practice controls.
What a patient types
The queries are short and they are local. A treatment word and a place word: lip filler Charlotte. A treatment word and “near me”. Sometimes the person types the problem instead, the lines between the eyebrows, and lets the results name the treatment. And increasingly the query is a sentence, typed or spoken to an assistant, asking who in Charlotte is good at this.
Google reads intent off the shape of the query. A treatment plus a place, or “near me”, is local intent, and for local intent Google leads with the map. A question gets a page, or, more and more, Google’s own AI summary above the pages. A sentence handed to an assistant gets read whole and answered.
The practice cannot change what people type. What it can do is make sure that each route, when asked, has something true and readable to say about it. That is the whole game, and everything below is a version of it.
The map
The map is the block of pins and cards at the top of a local search, and the first thing to know about it is that it is not built from websites. It is built from Google Business Profiles, the listings a business claims and edits directly with Google. A practice with a beautiful site and a neglected listing competes on the map with the listing.
Google chooses which listings to show, and in what order, on three things. Relevance: does the listing say it does this, in its primary category, its listed services, its name, and the words in its reviews. Distance: from where the phone is right now, which is why a practice sits differently in the map for a patient in Ballantyne than for one uptown. Prominence: how well known the practice appears to be, read off review count, how recently reviews arrive, the rating, mentions on other sites, and the authority of the website the listing links to.
The patient sees a card: name, rating, review count, category, open or closed right now, a photo, the distance. A tap opens the panel, with hours, reviews, a website link and, if the practice has set one, a booking link. A patient can go from the search to a booking page without ever visiting the website.
What the practice controls on this route: the category and the services on the listing, hours that are actually right, photos of the actual rooms and people, asking every patient for a review after every visit and answering the reviews that arrive, and where the website link and the booking link point. It does not control distance, and it does not control which handful of practices Google shows on a given afternoon.
Organic search
Below the map is the list, and the list runs on a different mechanism. Here Google reads pages, not listings. For a broad query such as med spa Charlotte, directories and roundup lists compete for the same slots as practices. For a treatment query, a practice’s own page about that treatment can appear, if it exists and says the right things.
What Google reads to decide: whether there is a page about this treatment at all, whether the words on the page match the words in the query, whether other sites link to it, whether it loads quickly and works on a phone, and whether the practice’s name, address and phone on the site match the listing. That last one is the quiet link between this route and the map. Google treats the site and the listing as one business when the facts agree, and as a puzzle when they do not.
Google’s AI summary now sits above the list for question-shaped queries. It is written from pages Google already ranks, so the page that answers the question in text is the page that feeds the summary. There is no separate trick.
What the practice controls on this route: a page for each treatment it offers, written in the words a patient uses, with the price or the range, what to do before and after, and a booking control on the page itself. A site that loads fast on a phone. Name, address and hours that match the listing exactly. Structured data under the page telling a machine what kind of business this is. It does not control the directories, and it cannot buy a place in the list.
AI answers
The third route is the newest and the least understood. A patient asks an assistant, in a sentence, who does lip filler in Charlotte and is good. The assistant has no directory of med spas. It either searches the web at that moment and reads a handful of pages, or draws on what it read while it was being trained, or both, and then writes a short answer that names a few practices.
The names come from pages the assistant can read: the practice’s own site, its listing, its reviews, directory entries, local press. It favours pages that state things plainly, in text, as facts it can quote: the practice’s name, where it is, what it does, what it charges. A page that says all of that in a sentence gets quoted. A page that says it in a hero image does not, because the assistant cannot read a photograph of words.
Two consequences follow. An assistant can only name a practice it can read, so a site that never renders its text for a crawler is invisible on this route. And an assistant hands the patient onward exactly as the map does, to a website or a booking page, and if the booking page sits behind a scheduling vendor’s robots file, the assistant cannot see it either. We wrote about that separately: that door is closed by the vendor, not the practice.
What the practice controls on this route: whether its site says, in plain text, what it does and where. Whether its name and details agree everywhere they appear. Whether it has reviews an assistant can find. Whether the page a machine reads is the page a patient reads. It does not control which assistant a patient uses, how that assistant searches this month, or whether it names anyone at all. The work we do on this route, the reviews, the listing and the site’s structure, is the inputs. The output belongs to the assistant.
Where the routes join
All three routes end at the same place, which is a tap. The booking button on the map panel. The booking button on the website. The link the assistant gives at the end of its answer. Every route delivers a stranger to that button, and what happens after the tap belongs to whichever scheduling vendor the practice uses.
That is why we spend most of our time on the website rather than on any single route. The site is what the listing links to, what the results rank, what the assistant reads and cites, and where the booking button lives. A practice found on all three routes that then shows a stranger a login form has done the difficult half and skipped the easy half. We walked the booking paths of med spa websites across the Carolinas in August and the findings are in the first article.
The websites we build are built for both ends of that job at once: a page for each treatment that a machine can read, facts that match the listing, and a booking control on every page.
What we cannot see
The honest part of this piece is the size of what it leaves out.
We cannot see the search logs. Nobody outside Google knows how many people in Charlotte searched for lip filler last month, or which route they took. Any split we gave you would be a guess dressed as a finding, which is why there is none.
We cannot see the route that never touches a search box. A friend’s face, a referral from a dermatologist, past patients coming back: word of mouth has no mechanism to walk through, and it is the route every practice would choose if it could choose only one.
We cannot see the paid route. Ads sit above the map and above the list, and a practice can buy them. We do not sell them, so this piece says nothing about them. That is a bias in what it covers, not evidence they do not work.
And we cannot see inside any of the three systems. Google does not publish how it weighs relevance against distance, the assistants do not publish which pages they read, and both change without notice. Everything above will need revising when the mechanism moves.
What this is not
This is not a promise about position. Nobody can guarantee a practice will show first on the map, in the list, or in an assistant’s answer, because each of those decisions is made by a machine the practice does not own, differently for every phone and every day. What can be promised is the inputs: the listing kept right, the reviews arriving, the site readable, the facts agreeing. That is a bet, and it is the honest one.
It is not a claim that Charlotte is different. The routes are the same in Greenville, Charleston and Columbia; what differs is who else is on the map. The practice’s task is the same everywhere: be readable, be accurate, be reviewed, and be bookable at the end of it. Everything we build is for practices across the Carolinas.
How to check your own
None of this needs us.
- Search for your own treatment from across town. On a phone, in a private window, somewhere that is not your practice. Type the treatment and the city. Note whether you are on the map and what the card says. Then open your listing as a stranger would and read the hours, the category and the latest reviews.
- Scroll past the map. If a directory outranks you for your own treatment, you do not have a page that answers the query, or the page does not say the words.
- Ask an assistant, in a sentence, who does that treatment in Charlotte. If it names you, ask where it got that. If it does not, look at your own site with the images turned off and ask whether there is anything left to read.
- Tap your own booking button, on each route. From the map panel, from the site, and from whatever the assistant linked. Count the screens before a real appointment time appears. If the count is more than one, that is where the work is.
If you would rather we did the last step for you, we will do that for free, and you keep the written report either way.