A real Foundry Audit, run on a real Orlando med spa — anonymized. Twenty sections, 78 documented findings, 27 prioritized actions. The first quarter is open below: the whole method, the scope disclosure, the score, and a sample from every section. The findings themselves are the deliverable clients pay for.
No email needed to read the preview. If the reasoning doesn't hold up, don't buy anything.
Meridian is a well-run practice with a data problem, not a quality problem. It holds a 5.0 rating across 214 Google reviews, the owner responds to reviews, the state license is clear with no discipline or complaints on file, and it ranks first organically for its signature device treatment. It is also close to invisible in the places AI answers are actually assembled from.
The split is unusually clean, and it is the most useful thing in this report. Where Google's own first-party data decides the answer — Maps, the local pack — Meridian performs well, appearing in both local queries tested and ranking second in its own neighborhood. Where the answer is assembled from third-party editorial and directory text — category listicles, review aggregators, vertical directories — Meridian appears in almost nothing. It was absent from all four "best med spa in Orlando" roundups examined, has no RealSelf presence, and its entries on three aggregators are unclaimed and carry review counts between 2 and 38 against a real count of 214.
That matters because AI assistants lean heavily on the second category. A business can own the map and still never be named in a generated answer.
Underneath that sit three compounding technical issues. The service pages — the pages that would answer
"where do I get X in Orlando" — carry no structured data at all. The homepage has no H1 and its
only Person markup names the web developer's admin account rather than the clinician. And the 214
reviews, the practice's single strongest asset, live inside a third-party iframe, meaning they are invisible to
any crawler or model reading the page.
None of this is exotic. All of it is fixable, and the highest-value fixes cost nothing but attention.
AnswerFoundry Visibility Score
A proprietary assessment, not a standardized industry metric. See §15 for the full category breakdown and how each score was reached.
What's already working
5.0 / 214 Google reviews with active owner responses. A Google Post published within the last week. A clear, active state license with no discipline or public complaints. First organic result for the signature device treatment. Consistent phone number across every source checked — the one clean identifier in the entire data set. This is a practice with real equity that is not being represented off-site.
An audit is only as good as the entity it tests. Before running a single prompt we establish what the business actually is, according to sources that can be verified independently of the business's own marketing.
| Attribute | Value | Source of truth |
|---|---|---|
| Business type | Independently owned med spa, single location | State corporate record; no chain affiliation found |
| Ownership model | Nurse-practitioner owned, operating under physician oversight | Practice's own About page |
| Lead practitioner | APRN, Nurse Practitioner qualification | Florida DOH MQA license verification, data as of 8 Aug 2026 |
| License status | Clear / Active · original issue 2020 · no discipline on file · no public complaint | Florida DOH MQA |
| Address of record | Suite 1, MetroWest, Orlando, FL 32835 | Florida DOH MQA — this is the authoritative address |
| Premises | A single suite inside a shared salon-suite building | Building operator's own directory |
| Google rating | 5.0 from 214 reviews | Google Maps, 8 Aug 2026 |
| Indexable pages | 32 URLs total — 8 site pages plus 24 service pages. No blog. | Site XML sitemaps |
| Published pricing | None on the practice website | Verified: zero occurrences of price indicators on service pages |
The tested entity is the business, not the website. This distinction drives most of what follows: AI systems resolve a business across many sources, and Meridian's website is only one of roughly a dozen places where its identity is asserted — several of which it does not control and has not claimed.
Scope disclosure — read before interpreting any result
This sample report was compiled without a client engagement. Testing covered the public retrieval substrate — the indexed sources AI systems draw on — and Google's local surfaces. Conversational AI assistants were not queried directly for this sample. No ChatGPT, Perplexity, Gemini, Claude, or Copilot session was run, and no finding below should be read as an AI assistant's answer.
In a paid Foundry Audit, ChatGPT, Perplexity, Claude and Google AI are queried, and every run is logged with its exact prompt text, timestamp, model version, full response and the sources cited — favorable or not. That log is the deliverable. This sample shows the structure and the reasoning; a client engagement fills the assistant rows with primary observation.
Not every engine is reached the same way, and the report says which is which. ChatGPT, Perplexity, Claude and Google Gemini are queried through their APIs with web search enabled — same retrieval substrate and citation behavior, but no session history or personalization. Google AI Overviews is captured from the live results page with its own cited references. Bing Copilot is not covered — Microsoft publishes no API for it and Azure's Bing-grounding tool withholds its output from developers, so any number we gave you would be a guess. Your report says so rather than quietly omitting it. Full methodology.
| Surface | Tested here | Method |
|---|---|---|
| Web search index | Yes | 10 queries, US locale, 8 Aug 2026 — full result sets recorded |
| Google Maps / local pack | Yes | 2 location-scoped queries, browser, 8 Aug 2026 |
| Third-party directories & aggregators | Yes | 9 platforms inspected directly |
| Site structured data | Yes | JSON-LD extracted from the live rendered DOM, 4 page types |
| State licensure record | Yes | Florida DOH MQA license verification |
| ChatGPT | Not tested | Queried in a client engagement via the API with web search enabled, not in this public sample |
| Perplexity | Not tested | |
| Claude | Not tested | |
| Google Gemini | Not tested | |
| Google AI Overviews | Not tested | Captured from the live results page in a client engagement, with the overview's own cited references. A different surface from Gemini, so reported separately |
| Bing Copilot | Not tested | Not covered in the sample or in a paid audit. Microsoft publishes no API and Azure's Bing-grounding tool withholds its output from developers, so no claim is made. For your own site, Bing Webmaster Tools now reports Copilot citations directly — we will point you at it |
Prompts are written the way a prospective patient would ask, not the way a marketer would search. Three families are used, and the mix is fixed before testing begins so results cannot be curated after the fact.
| Family | What it measures | Count |
|---|---|---|
| Category discovery — "best med spa Orlando" | Whether the business is a candidate answer at all for high-intent, unbranded demand | 6 |
| Service-specific — "[treatment] Orlando FL" | Whether individual service pages surface independently of the brand | 2 |
| Adjacent / qualifier — ownership model, ZIP, neighborhood | Whether the business is retrievable when a customer adds a constraint | 1 |
| Branded control — "[business name] reviews" | Baseline: confirms the business is retrievable at all, isolating a discovery failure from an indexing failure | 1 |
Share of answer is the percentage of relevant, unbranded queries in which the business appears as a candidate answer. It is the headline number, and on its own it is close to useless without the detail beneath it — which is why the full query log follows rather than a single figure.
Ten prompts were run: nine unbranded and one branded control. One of the nine — query 4 — is a legitimate non-fit and is excluded from the denominator rather than counted as a miss, which would overstate the problem. Share of answer is therefore scored across 8 queries. Query 6 counts toward the numerator even though Meridian surfaced only through a third party showing incorrect data, because the customer would still have seen the business named.
Visual 1 — Platform visibility table
| # | Query (exact text, 8 Aug 2026) | Appeared? | How / notes |
|---|---|---|---|
| 1 | best med spa Orlando | No | 7 competitors named; 4 of 9 result domains were third-party listicles |
| 2 | best med spa near me Orlando FL | No | 9 competitors named |
| 3 | top rated botox Orlando | No | An aggregator holding an unclaimed Meridian profile took the top two result slots — and the profile never surfaced |
| 4 | physician owned med spa Orlando | N/A | Correct non-fit. Meridian is NP-owned. Included to map the competitive frame, excluded from scoring |
| 5 | best place for filler Orlando FL | No | 9 competitors named |
| 6 | med spa MetroWest Orlando | Partial | Named first — but sourced from a review platform that displays Meridian at 3.0 from 2 reviews. Its own domain did not rank |
| 7 | [signature device] Orlando FL | Yes — #1 | Own service page, first organic result. Low-competition device term |
| 8 | laser tattoo removal Orlando FL best | No | Notable: tattoo removal is Meridian's most-reviewed service on Google, with 53 review mentions |
| 9 | "[business name]" med spa Orlando reviews | Yes | Branded control. Dominant, as expected — this is an indexing check, not a discovery result |
| 10 | nurse practitioner owned med spa Orlando 32835 injectables | No | Absent even with its own ZIP code and its own ownership model in the query |
Query 10 is the most diagnostic line in the table. A prospective patient who already knows they want an NP-owned injectables practice in Meridian's exact ZIP code would not be shown Meridian. That is not a ranking problem; it is an absence-of-published-text problem, and §12 explains why.
Visual 2 — Local surfaces, for contrast
| Local query | Result |
|---|---|
| med spa near MetroWest Orlando FL (Maps) | Ranked #2 in the local feed — 5.0 (214), correct address |
| best med spa Orlando FL (Maps) | Appeared in the local feed — 5.0 (214) |
Two tables, one business, opposite outcomes. Meridian's Google Business Profile is doing its job. Everything downstream of Google's own data is not.
Everything above is method and measurement: how the testing was run, what was and wasn’t tested, and where this business actually shows up. That part is free on purpose — it’s the part you should be able to argue with before spending anything.
What follows is the diagnosis: specific, named, dated defects, and the order to fix them in. Every section below stays open so you can see its shape, but shows only its first two rows.
A Foundry Audit is this report, run on your business, with each engine queried and every run logged — exact prompt, timestamp, model version, full response and the sources cited. Delivered in about two weeks, and credited toward your first month of implementation if you start within 14 days. No placement, ranking or citation in any AI answer is guaranteed — see §20. Full pricing and terms.
Competitors are anonymized and characterized only by observable, verifiable attributes. Google review counts were read directly from Maps on 8 August 2026. No judgment is offered about any competitor's quality of care.
Visual 3 — Competitor comparison
| Google reviews | Queries appeared in | Observable advantage over Meridian | |
|---|---|---|---|
| Meridian (subject) | 5.0 / 214 | 2 of 8 | — |
| Competitor A | 4.9 / 525 | 4 | Named physician on every page; ~60+ indexed treatment pages; publishes its own "Top 10 Orlando Med Spas" listicle that ranks for the head term and places itself first |
| 4 more competitor profiles withheld. Each names the observable advantage that competitor holds — including the one that publishes its own “Top 10 Orlando Med Spas” list and places itself first. Unlock → | |||
Every row below is a data-consistency observation with a named source and an observation date. None of it is a characterization of the business.
Visual 4 — Accuracy issue table
| # | What's published | Where | Correct value | Sev. |
|---|---|---|---|---|
| 1 | 3.0 stars from 2 reviews | Major review platform — and this is the figure its neighborhood listicle propagates | 5.0 from 214 (Google) | High |
| 2 | "5.0 from 180+ reviews" | Meridian's own site footer, every page | 214 | Med |
| 13 more accuracy defects withheld — 6 rated High or Critical. Each pairs what is published against the correct value, with a severity rating. Two of the withheld rows carry regulatory weight in a licensed profession. Unlock → | ||||
This section asks a narrow question: if a customer described what they wanted in plain language, is there text anywhere that would let a machine match them to this business?
| Attribute a customer might specify | True of Meridian? | Stated in a title, heading, or schema field? |
|---|---|---|
| Nurse-practitioner owned / NP-led | Yes | No — body copy on one page only |
| Operates under physician oversight | Yes | No — body copy only |
| 6 more attribute gaps withheld. Every withheld row is a true fact about this business that it is simply not publishing anywhere a machine can read it. Unlock → | ||
This is the section most visibility reports skip, and the one that explains the results. Rather than asking "why isn't the business ranking," it asks: which sources decide this category's answers, and is the business present in them?
| Source type | Appeared in | Meridian's status |
|---|---|---|
| Review-platform category and neighborhood pages | 7 of 10 queries, often multiple slots | Present but wrong — 3.0 / 2 reviews, wrong suite, name casing error. The most influential source in the category, showing the least flattering data. |
| Competitor-authored "Top 10 / Best of" posts | 4 queries | Absent from all |
| 4 more source types withheld. This is the section most visibility reports skip, and the one that explains the results — which sources decide this category’s answers, and whether the business appears in them. Unlock → | ||
All findings extracted from the live rendered DOM on 8 August 2026, not inferred from a crawler summary.
| Finding | Detail | Impact |
|---|---|---|
| Service pages carry no structured data | Zero JSON-LD blocks on the service page inspected. No Service, FAQPage, Offer, MedicalProcedure, or Product. These are the pages that answer treatment queries. | High |
| Reviews are invisible to machines | The reviews page renders ~555 characters of text; all review content sits in a third-party iframe. No Review or AggregateRating markup anywhere on the site. | High |
| 8 more technical findings withheld — 4 rated High or Critical. All read from the live rendered DOM on the test date, not inferred from a crawler summary. Unlock → | ||
Entity consistency is how confidently a machine can conclude that scattered mentions refer to one business. Low confidence produces hedged answers, or omission.
| Identifier | Variants live | Assessment |
|---|---|---|
| Suite number | 4 — Suite 1 (state record), Ste 2, Suite 20, none | Critical |
| Business name | 6 — spacing, casing, and LLC-suffix variants | Critical |
| 7 more identity conflicts withheld — 4 rated High or Critical. How many variants of each identifier are live simultaneously, and the single likely root cause behind the whole pattern. Unlock → | ||
| Gap | Observed | Competitor benchmark |
|---|---|---|
| Total indexable pages | 32 | Competitor A: ~60+, including a treatment blog cluster |
| Blog / editorial | None | Two competitors publish category roundups that rank for the head term |
| 6 more content gaps withheld. Each set against what the competitive set actually publishes — plus the three faster moves that come before “publish more.” Unlock → | ||
| Signal | Status |
|---|---|
| Google Business Profile — claimed and active | Strong. Correct address, correct hours, owner responses, a Post within the last week |
| Review volume and rating | Strong. 5.0 / 214 — 6th largest count in the local competitive set |
| 8 more local signals withheld. The one strong asset, the neglected surfaces around it, and why the strong asset isn’t converting into answers. Unlock → | |
Observations from public sources, not legal or regulatory advice. Anything below with regulatory implications should be reviewed by the practice's own counsel or compliance advisor.
| Observation | Why it matters | Priority |
|---|---|---|
| A third-party page describes the practitioner as a physician | She holds a clear, active APRN license. Neither the practice nor its agents authored this text, but uncorrected public misstatement of a clinical credential is a regulatory and trust exposure in a licensed profession | High |
| A third-party profile publishes prices the practice does not publish itself | Prices quoted by a source the practice does not control may be stale. In aesthetics, advertised pricing carries consumer-protection implications | Med |
| 4 more risk observations withheld — 1 rated High or Critical. Public-source observations with regulatory or consumer-protection implications, each with a priority. Unlock → | ||
This is AnswerFoundry's proprietary assessment methodology. It is a structured way to compare a business against itself over time and against observable competitor attributes. It is not a standardized industry metric, it is not calibrated against an external benchmark, and it should not be read as implying scientific precision. The reasoning for each category is shown so the number can be argued with.
AnswerFoundry Visibility Score
The total is shown. The five category scores behind it — entity accuracy, website technical readiness, content coverage, authority and citations, local presence — and the written reasoning for each are part of the paid deliverable, because the reasoning is the part worth arguing with.
| Category | Score | Reasoning |
|---|---|---|
| All 5 category scores and their reasoning withheld, along with the effort/impact priority matrix that sorts every finding into do-now, plan, fill-in and later. Unlock → | ||
The score is a structured way to compare a business against itself over time, and against observable competitor attributes. It is not a standardized industry metric, is not calibrated against an external benchmark, and should not be read as implying scientific precision.
Objective: stop publishing contradictory information about yourself, and put the head terms where machines look.
| # | Action | Owner | Effort |
|---|---|---|---|
| 1 | Reconcile hours across the site's visible text, its structured data, and the Google Business Profile. Pick one truth and publish it in all three | Practice + web | 1 hr |
| 10 more 30-day actions withheld, each with an owner and an hour estimate. Unlock → | |||
Objective: make the existing assets machine-readable, and claim the surfaces that feed generated answers.
| # | Action | Owner | Effort |
|---|---|---|---|
| 10 more 60-day actions withheld, each with an owner and an hour estimate. Unlock → | |||
Objective: earn presence in the third-party sources that decide this category's answers.
| # | Action | Owner | Effort |
|---|---|---|---|
| 6 more 90-day actions withheld, each with an owner and an hour estimate. Unlock → | |||
Every prompt run — in the baseline and in every cycle after — records the following fields. All of them, every time, whether the result is favorable or not.
| Field | Why it's recorded |
|---|---|
| Date and time | Answers change; an undated result is not evidence |
| Platform and, where visible, model version | Results are not comparable across platforms |
| Exact prompt text | Paraphrasing between cycles destroys comparability |
| 9 more logged fields withheld, plus the re-test cadence, the reporting format, and what is deliberately not promised as a metric. Unlock → | |
A structured record of what publicly available sources said about one business on one day, and a professional assessment of the signals most likely influencing that. It is a diagnosis and a plan, not a prediction.
No-guarantee statement
AnswerFoundry does not control ChatGPT, Google, Bing, Perplexity, Gemini, Claude, or any other third-party AI or search system, and does not guarantee placement, citation, ranking, recommendation, or mention in any AI-generated answer. No representation is made that any specific result, position, traffic level, lead volume, or revenue outcome will be achieved. Recommendations reflect observed data and professional judgment as of the test date. Client results vary. Historical visibility does not indicate future visibility. Some work described here may also affect traditional search performance, but AI visibility and traditional SEO are not the same thing and should not be expected to move together.
This report contains no legal, medical, regulatory, or financial advice. Observations touching on professional licensure or advertising should be reviewed by the practice's own counsel or compliance advisor.
The subject is a real, independently owned Orlando-area med spa, selected without its knowledge or participation from public information only. Its name, address, and practitioner identity have been changed, and competitors are identified by letter. No confidential information was used, and no source was accessed other than pages available to any member of the public. All findings are data-consistency observations with a named source type and an observation date; none constitutes a characterization of the business, its practitioner, or any competitor. Every business in this report — subject and competitors alike — appears to be a legitimately operated practice.
Any business that recognizes itself here may contact hello@answerfoundry.ai and this report will be amended or withdrawn on request, without argument.
A Foundry Audit is this report, run on your business, with every AI assistant queried directly and logged — exact prompts, dates, full responses, screenshots. Every result, including the ones that don't flatter anyone. You get all of it, not 23% of it.
Foundry Audit — $795 standard; $495 founding launch. The launch price ends September 19, 2026 or after the first 10 paid audits, whichever comes first. Credited toward your first month of implementation if you start within 14 days.
Claim the founding audit — $495 Start with the free Answer Snapshot
Not sure yet? The free Answer Snapshot is a short version of sections 5, 6 and 7 run on your own business — enough to tell you whether there's a real gap worth paying to investigate. See full pricing and terms.