Why does ChatGPT say my optometry clinic doesn't take a patient's vision plan when we do?

Almost always it is reading a stale source or no source at all. An old directory might still list a plan you dropped, or there may be no machine-readable insurance list anywhere, which makes the AI guess and often guess wrong. Eye care has an extra trap: routine exams and glasses usually run through a vision plan while medical eye visits run through regular medical insurance, so an AI can decide you only do glasses. The fix is to write your current vision plans by name in your Google Business Profile and in plain text on your site, note that you bill medical insurance where you do, and check that older directories are not contradicting you. Wrong insurance is one of the most damaging errors for a clinic, because it turns away the exact patients searching for you.

Customers ask the AI for an eye exam this week. How do I show up for that?

Two things decide it. Your hours have to be current and accurate, including any evening or Saturday hours, so a wrong "closed" flag does not drop you from the availability searches. And your "accepting new patients" signal has to be turned on in your profile and stated in words on your site, ideally with a line about how soon a new patient can usually be seen. A clinic that reads as open, taking new patients, and quick to book is the one a same-week query lands on, because the AI is matching the urgency in the question.

We're a small clinic with a few dozen reviews. Can the AI name us over a big chain?

Yes. Bigger is not the signal, clearer and fresher is. A clinic with thirty recent reviews that mention a thorough exam and no pressure to buy glasses, an accurate vision-plan list, real hours and an "accepting new patients" flag can beat a large chain with old data and a conflicting address. Because eye care is a health trade, the AI leans toward the clinic whose details it can verify, and a small clinic that answers the patient's exact question precisely often wins that specific query.

How should I ask patients for reviews without breaking the rules?

Ask every patient, not just the ones who seemed delighted, since picking out the happy ones or screening out the unhappy ones breaks Google's policy. Make it easy, like a text with a review link the same day, and invite the patient to describe their visit in their own words. Do not tell them what to say, do not coach phrases like "thorough" or "worth it", and never offer a discount or a freebie for a review, which crosses Google and FTC rules. When patients naturally mention the parts that matter, like the careful exam or feeling no pressure to buy glasses, the AI reads those words on its own, which is exactly the point.

How to get your optometry clinic recommended by ChatGPT and AI

Someone in your town breaks their glasses, or their kid fails the school vision screening, and instead of opening Google they open ChatGPT and type \"eye doctor near me that takes VSP and can see my son this week.\" The AI thinks for a second and gives back two or three clinic names with a line on each. Your clinic is either in that short answer or it is not. There is no second page to scroll. The decision is made in the first three names. The useful part is that the AI is not inventing those names. It reads a small set of public sources about your clinic and repeats what it finds, and most of those sources are things you control. Because eye care is a health trade, these AIs are careful: they favor clinics whose details they can verify and go quiet on the ones they are unsure about, so accurate, readable information wins more here than almost anywhere else. This playbook walks through what ChatGPT, Gemini, Perplexity and Google AI Overviews actually read about an optometry clinic, where they get your details wrong, and the handful of fixes that decide whether a parent or a patient with red, watery eyes hears your name.

The questions patients actually ask the AI

People do not ask an AI the way they type a keyword into a search box. They ask in full sentences, the way they would ask a coworker who wears glasses, and they pile on conditions. An eye-care query almost always carries a constraint about insurance, about who the patient is, or about how soon they can be seen.

These are the shapes that come up over and over for optometry clinics. Notice how often a plan name or a specific service is baked right into the question.

"Eye doctor near me" and "eye exam [city]" for general searching
"Optometrist that takes VSP" or "eye doctor that accepts EyeMed near me", with the vision plan named up front
"Kids eye exam near me" and "pediatric optometrist that's good with young children", where the worry is the experience, not the prescription
"Contact lens fitting [city]" and "eye doctor that does first-time contact lens fittings", a specific service as the whole query
"Eye doctor for dry eyes near me" and "LASIK consultation [city]", where a medical concern or a procedure drives the search
"Eye exam this week near me" and "optometrist accepting new patients", where speed and availability are the point

What the AI reads about an optometry clinic

When the question is local, the AI leans hardest on your Google Business Profile, then on your reviews, then on your own website to confirm what it found. Google's AI Overviews lean heavily on that profile too for local results. For an eye clinic, a few of those fields carry far more weight than the rest, because they answer the exact constraints patients attach to the question.

Accepted vision insurance is the field that gets typed into the query most and is the hardest for the AI to find on its own. Vision plans like VSP, EyeMed, Davis Vision and Spectera are not interchangeable, and a patient who has VSP wants to hear that you take VSP, by name. There is a wrinkle unique to eye care: a routine exam and glasses usually run through a vision plan, while a medical visit for dry eye, pink eye or a diabetic eye exam runs through regular medical insurance. If your profile and site do not make both clear, the AI cannot match you to either kind of question and tends to skip you.

Your services are the next priority, and they have to be named one by one. A comprehensive exam is assumed, but contact lens fittings, on-site glasses and an optical shop, pediatric exams, dry eye treatment, and LASIK co-management each need to be stated explicitly or the AI will not connect you to a query that asks for them. Same-week availability and a clear new-patient signal decide whether you appear for the ready-to-book searcher. The fields that move the needle:

Accepted vision plans written out by name (VSP, EyeMed, Davis Vision, Spectera, and so on), plus a clear note that you also bill medical insurance for medical eye visits where you actually do
Services spelled out individually: comprehensive eye exams, contact lens fittings, glasses and on-site optical, pediatric exams, dry eye treatment, and LASIK or surgical co-management
A clear "accepting new patients" signal in the profile and stated in words on your site
Real hours, including any evening or Saturday hours, with holiday hours kept current
The correct primary category (optometrist) rather than a vague "eye care" or "glasses store" label, so you match the right kind of query
Name, address and phone that match your website letter for letter, since a moved clinic or a second location splits the picture fast

The wrong facts that cost an optometry clinic the most

Because eye care is a health trade, an AI that is unsure about your clinic tends to say less, hedge, or move on to a clinic whose details it can verify. That makes a single wrong fact more expensive for you than it would be for a coffee shop. Two errors do the real damage, and they are the two patients screen on hardest.

The first is wrong insurance. If an old directory still lists you as taking a vision plan you dropped, or the AI cannot find your plans at all and fills the gap with a guess, you get two bad outcomes at once: patients call about coverage you do not have, and patients who do have the right plan never hear about you. The vision-versus-medical split makes this worse, because an AI can decide you only do glasses and quietly leave you out of "eye doctor for dry eyes" answers. The second is the new-patient problem. If anything in your public footprint reads as not taking new patients when you are, the AI stops recommending you for the exact query you most want to win, the patient ready to book an exam this week.

Then there are the basics that sink any local business but land harder on a clinic patients are trusting with their eyes:

A vision plan listed that you no longer accept, or no readable insurance list at all, which makes a cautious health-aware AI guess or skip you
Missing the medical-billing side, so the AI files you under glasses only and drops you from dry eye, pink eye and diabetic exam questions
A stale "closed" flag left over from a move, a remodel or a change of ownership, which removes you from every recommendation while you sit fully booked
Old hours that say you close at 5pm when you now stay open until 7pm, or omit your Saturday hours, so you lose the after-work and weekend searcher
A wrong phone number, which for a patient with an irritated eye just means they call the next name on the list
Conflicting name, address or phone across your site, Google and directories, which makes the AI hedge instead of name you

Reviews, and the themes the AI pulls out of them

Between two clinics that look the same on paper, the AI almost always names the one with more reviews and a higher, fresher rating. Review count and recency read as "this place is real and patients keep coming back." You cannot fake that and should not try. What works is a steady, honest habit, not a one-time burst.

The rule that keeps you safe and also works best is simple: ask every patient for a review, not only the ones who seemed thrilled. Picking out the happy ones to ask, or holding back the ask from anyone unhappy, breaks Google's policy and can get your reviews pulled. Invite the patient to describe their visit in their own words. Do not hand them lines to repeat, do not tell them to say it was worth it, and never offer a discount, a free fitting or any incentive for a review. All of that crosses Google and FTC rules. The honest version is enough.

For eye care, what the reviews say matters as much as the count, because patients are screening for a particular kind of experience. The AI reads review text and pulls out recurring themes, and when patients naturally use the same words, the AI can surface you for the query those words match. You do not script this, you just make the experience worth describing. The themes that tend to decide an eye-care recommendation:

"Thorough exam" and "didn't feel rushed", which a careful, health-aware AI reads as a clinic that does the full job
"No pressure to buy glasses", a real fear for patients and a strong differentiator when it shows up on its own in reviews
"Great with kids" and "my daughter wasn't nervous", which feed the "kids eye exam near me" question
Mentions of a specific service done well, like a first-time contact lens fitting or finally getting help with dry eye
"Easy to get in" and "saw me the same week", which line up with the availability searches you want to win

The two or three fixes worth doing first

Most owners try to fix everything at once and finish nothing. For an optometry clinic, a short list gives you the most movement for the least effort.

First, write your accepted insurance somewhere a machine can read it. List your vision plans by name (VSP, EyeMed, Davis Vision, Spectera, whichever you take) in your Google Business Profile and in plain text on your website, not buried in a PDF or trapped inside an image. If you also bill medical insurance for medical eye visits, say so plainly, because that one line is what lets you appear for dry eye and pink eye questions instead of glasses-only ones. Only list plans you genuinely accept and services you are actually licensed and equipped to provide; an AI that is later contradicted by a patient stops trusting your listing, and overstating what you do is the wrong kind of attention.

Second, set and confirm your new-patient status and your real hours, including any evening or Saturday hours, so the highest-intent searcher, the one who wants an exam this week, actually finds you. Third, start the review habit and ask every patient, in a neutral way, for an honest review. A few concrete moves:

List every vision plan you take, by name, on your profile and on a plain-text page of your site, and note medical billing where it applies
Name each service on its own (exams, contact lens fittings, on-site optical, pediatric exams, dry eye treatment, LASIK co-management) so the AI can match each query
Turn on the "accepting new patients" signal and say it in words on your homepage, alongside how soon a new patient can usually be seen
Add LocalBusiness structured data to your site, a small block of code that states your name, address, phone, hours and services in a format the AI reads without guessing
Text a review link to every patient after the visit, ask for an honest account in their own words, and when you reply, keep it genuine rather than coaching the next reviewer

Then check where you actually stand

After you fix the insurance list and the new-patient signal, the step almost every clinic skips is confirming whether the AI changed its answer. Fixing your inputs and never checking the output leaves you guessing about whether any of it worked.

The honest way to check is to ask the AIs the questions a patient would ask, in your own city: the insurance one, the kids one, the contact lens one, the "eye exam this week" one. Ask each a few times, because the answers shift from run to run, so a single check tells you almost nothing. What tells you something is your mention rate, how many of those repeated runs across ChatGPT, Gemini and Perplexity actually named your clinic. If you go from zero out of nine to five out of nine after fixing your vision-plan list and your new-patient flag, you know the fix landed. It also catches the thing you cannot see from the exam room: the exact wrong fact an AI is repeating about you, quoted word for word, and the competing clinic it keeps recommending in your place.

That is the part LocalFox does for you. The free check gives you an AI visibility score and your single biggest problem, with no account. The $9 report runs each of those patient questions three times across ChatGPT, Gemini, Perplexity and Google AI Overviews, quotes back word for word what they say about you, including a wrong vision plan or a false "closed" flag, shows which competing clinics they recommend instead and the reasons they give, and hands you a copy-paste fix kit: review-request templates, a Google Business Profile description draft, and a LocalBusiness schema block. It is a one-time $9 report, not a subscription and not a dashboard, with no card kept on file, and it includes one free re-scan within 60 days so you can confirm your fixes worked. Nobody can promise placement, because there is no ad slot inside an AI recommendation, but you can see exactly what the AI says about your clinic today and fix the inputs it reads.