For concierge practices
Is your concierge practice found in ChatGPT, Gemini, Perplexity, Claude, DeepSeek or Mistral?
1 September 2026 - we asked an assistant
“concierge doctor in Austin, TX”
Your customers now ask ChatGPT, Gemini, Perplexity, Claude, DeepSeek and Mistral where to go for a concierge doctor. Do you do enough to get chosen?
What gets observed
ChatGPT chooses others. What about you?
Real local questions, asked on the interfaces customers actually use. Watch who gets named - and who doesn’t.
What we see for concierge practices
AI already answers this question every day. From a list of five or six names.
These are the kinds of questions we measure for concierge practices:
When someone puts a local question like these to an assistant, it usually answers with a short business list. Across the categories we have measured, that list is drawn from a pool of five or six businesses - four or five of them shown at a time. Ask again and the pool holds while the line-up moves: of 253 names we saw across 45 questions, 119 came back in every reading and 134 did not. In 42 of those 45 questions at least one name never dropped out. The free scan shows whether you are in the pool - and if you are, whether you are one of the names that holds. And it is your city, not your industry. We asked the same question in 3 metros across 10 trades: of 19 comparable pairs of cities, not one shared a single business. The short list you are competing for is rebuilt for every city. The list is not guaranteed to appear at all: in 45 of 63 readings the assistant rendered one, and in the other 18 it named businesses in prose instead. Every figure above describes the list. The list is not the whole answer. In 53 of the 103 readings that showed a business list, the text beneath it named businesses the list did not carry; and in 53 of the 58 readings that showed no list at all, the text still named some. We read both.
We measured concierge practices in Austin, TX directly: one local question, asked twice. Across those readings the assistant drew on 4 concierge practices in all, and every one of them came back every single time. The same practice came first in every one of those readings. We also recorded what the assistant read to build its answers about concierge practices: of 14 citations, 14 pointed at a business's own website. For concierge practices, your own pages are the surface it reads. The list is not the whole answer: in 2 of the 3 readings that showed a list, the text beneath it named concierge practices the list did not carry.
Persona Healthcare - Concierge internal medicine with $250/month membership, direct physician access, extended appointments, and same/next-day appointments.…
Quoted word for word from the answer we stored. The first business the answer named, in its own words.
How it works
Our AI Agent asks. You get the answer and the fix.
The questions your customers put to AI assistants, asked on a schedule, recorded word for word, and turned into a report and a fix list.
-
1Step 1 of 3
concierge doctor near me
direct primary care monthly cost in Tampa
dpc doctor taking new patients
It asks what your clients ask
The questions someone looking for a concierge doctor in your area asks - put to five AI assistants, again and again.
-
2Step 2 of 3
Named in AI answers 32%
Named in 32% of the answers
You get the report
Who was named, who was not, and which pages decided it.
-
3Step 3 of 3
Named in AI answers
29.4%+42.0 pp- Answers checked1,817+13.4%
- Named you534+551%
Day Answers checked Named you 7 Aug 52 3 8 Aug 58 4 9 Aug 49 4 10 Aug 61 5 11 Aug 55 5 12 Aug 63 7 13 Aug 57 7 14 Aug 60 8 15 Aug 54 9 16 Aug 66 11 17 Aug 59 11 18 Aug 62 13 19 Aug 56 13 20 Aug 64 15 21 Aug 58 16 22 Aug 61 17 23 Aug 65 19 24 Aug 57 19 25 Aug 63 21 26 Aug 60 23 27 Aug 68 24 28 Aug 62 26 29 Aug 59 26 30 Aug 66 28 31 Aug 64 30 1 Sep 61 31 2 Sep 67 33 3 Sep 63 34 4 Sep 65 35 5 Sep 62 37 Fix what our report points at
A short list of fixes, each tied to its answer. Or we do them for you.
Around the clock
We analyze data 24/7
Our agent keeps putting your customers’ questions to the assistants - day and night - and every answer lands in your record.
The report
See if your business is found by AI. Then decide how far to go.
Start free. If it is worth going further, the report is one payment - and every price here is what you pay.
Scroll down for monitoring and the site work ↓
Free scan
$0
No card
- We ask AI one question your customers ask
- You see the three names it gives back
- And whether yours is one of them
Answer by email. No card, no call.
The report
$99
One time
- About 60 questions your customers really ask
- Six AI services, ~360 answers, every one saved word for word
- Who gets named instead of you, and why
- Your site read page by page, with the fixes
- A ranked plan, in your inbox within the hour
One payment. No subscription.
Monitoring
Are you still in the answer?
One reading is a snapshot of something that moves. Monitoring asks the same questions on a schedule and shows you who moved in or out.
Monitoring
$79/mo$99
Billed monthly, with our badge
With our badge in your footer. Without it, $99/mo.
- We ask again, on a schedule
- You see if you are still in the answer
- And who moved in or out
The report answers where you stand today. Monitoring answers whether you held - and nothing else does: 134 of the 253 names we saw did not survive a re-read.
Cancel any time. First month free with any report.
Invite a friend
$0/mo$79
One free month per paying referral
- Send your link to a business owner you know
- They run a free scan - free for them, nothing owed
- They buy their first report, and your next month is free
The month comes off your next invoice. Your badge has to be up on the day they pay.
The next step
Do you want us to do it for you?
We use your own results, not a template. Pay first, we start. Limited slots.
Fix my website
$399$499
One time, with our badge
With our badge in your footer. Without it, $499.
- We start from your report, so the work aims at the questions that failed
- A list, page by page, of what an assistant cannot read on your site
- We make the changes on your site ourselves, once you have approved them
- The same questions run again afterwards, and reported either way
Pay first, we start. Limited slots.
Build a new website
$799$999
One time, with our badge
With our badge in your footer. Without it, $999.
- Built from your measurement, aimed at the questions the old site failed
- Every fact a customer needs is text on the page, not locked in a widget
- One page per service, written to be quoted
- Measured after launch and reported, either way
Pay first, we start. Limited slots.
What decides it
Which practice gets named? Three things decide.
We measure how AI builds its “best near me” answers. The pattern holds - and it is not your Maps ranking.
The Google card
Assistants pull their local candidates from the same data Google Maps runs on. Your Business Profile gets you considered - or leaves you invisible before anything else matters.
The review filter
In the categories we’ve measured, the businesses AI names carry review counts several times their category’s median. Reviews decide who survives the cut.
Your own website
Most of what AI actually says - prices, who does same-day access best, what to expect - it quotes from somewhere. For concierge practices we measured that it is the businesses' own sites, and of the 34 businesses ChatGPT named that we could match to our Maps sample, 34 run a live site of their own. Having a site isn't the bar - having one AI can quote is.
Maps position doesn’t explain it. In our measurements, businesses ranked past #60 in Maps got named while some top-ranked ones were skipped. The baseline shows which of the three factors above is failing you - with the evidence attached.
FAQ
Before you start.
Can’t I just ask ChatGPT myself?
You can, and you’ll learn something. But one answer on one day from one service isn’t a measurement. We check several services, repeat the questions people actually ask, and keep the raw answers so the conclusion can be checked.
Does this replace our SEO or our marketing agency?
No. It measures a surface most agencies don’t report on yet. The output is evidence you can hand to whoever already does your marketing.
What do you need from us?
Your website, your city, and an email address. Nothing confidential, nothing about your customers.
Our panel fills by referral - why measure at all?
Panels churn, and the patients replacing them increasingly start with an assistant, not a neighbor. If your panel is full, the result still tells you what your waitlist pipeline looks like - and whether the names AI gives instead are your future competitors.
Is the monitoring subscription required?
No - the scan, the report and the fix work are one-time purchases and stand on their own. But one reading is a snapshot of something that moves: most of the names we have measured did not come back unchanged on a re-read. Monitoring is what turns a single reading into a position you can watch. Any report purchase includes the first month free.
What if the scan shows we’re doing fine?
Then you’ve spent nothing and you know it. The free scan exists to tell you whether the paid one is worth it - sometimes the answer is no.
Make your concierge practice seen in ChatGPT, Gemini, Perplexity, Claude, DeepSeek and Mistral.
One minute, no card, private result.
Step one
See where your practice stands. Free.
Four short answers. No card. We use your business name, your website and your city - nothing else.
- One free scan per email address
- Public business information only
- Result goes to you privately, not to a dashboard anyone can see
Free scan
Takes about a minute.