What Patients Ask AI Before Choosing a Practice

There is a stretch of a patient's decision you never see. It happens weeks before the phone rings, in a private chat with an AI, where the patient is trying to talk themselves into, or out of, a treatment. They are not yet looking for you. They are asking whether the thing is worth doing at all. This is documented behavior, not a projection. KFF found that about a third (32%) of adults nationally said they turned to AI chatbots in the past year for health information (KFF, 2026), and 16% used AI specifically to help decide whether to see a doctor, with 19% using it to understand and compare treatment options (KFF, 2026). For a cosmetic dental, orthodontic, or aesthetic practice, the live question is simple: when that private conversation names a provider or a resource, is it yours? This page maps the exact questions patients ask, using our first-party reading of real patient searches, and what that means for the practice that wants to be in the answer.

Patients Are Already Asking AI About Their Care

The shift is not coming, it is here. KFF found that 16% of adults used AI in the past year to help decide whether to see a doctor, and 19% used it to understand and compare treatment options (KFF, 2026). That is the exact slice of behavior that decides whether a high-consideration procedure happens at all. And it is not a fringe habit: KFF found that about one in six (17%) adults now use such chatbots at least once a month to find health information and advice, including a quarter (25%) of those under age 30 (KFF, 2024). Elective cosmetic and orthodontic patients skew toward exactly that younger, research-heavy profile.

Not every patient does this, and overstating it would be a mistake. But enough do, and the trend is toward more, that ignoring the research phase means ceding it. Worse, even when the patient does turn to a regular search, the answer now often arrives before any click. Pew Research, using real browsing data, found that 18% of Google searches produced an AI summary, and that when one appeared, users clicked a result just 8% of the time versus 15% when none appeared (Pew Research Center, 2025). The patient gets their answer. The question is whose content supplied it.

The Actual Questions They Type

This is where first-party data beats theory. Cakesmash Signal Mining of Google Autocomplete (June 2026) captured hundreds of real patient searches across the three niches we serve, and the cosmetic dental category was the richest. The questions are not "cosmetic dentist near me." They are evaluative: "are veneers worth it," "are dental implants worth it," "is teeth whitening worth it." The patient is still deciding whether to do the thing. Then comes comparison: "should i get veneers or bonding," "are implants better than dentures," "are implants better than a bridge." Only later does the search turn to a provider: "how to choose a cosmetic dentist," "best cosmetic dentist near me," "how to choose a dentist for implants."

Cost sits woven through all of it: "how much do veneers cost," "are veneers worth the money," "why are dental implants so expensive reddit." Notice the shape of these. A patient typing "are veneers worth the money" is not asking for a price, they are asking for a reason to spend it. AI answers built from content that lists a price range with no rationale miss the question entirely. The patient wanted longevity, candidacy, and a comparison to the alternative they will regret skipping. The practice whose content supplies that becomes the citable answer. The practice with a services page and a gallery does not.

The Worth-It and Reddit Pattern That Dominates Every Niche

The single most important finding from Cakesmash Signal Mining (Google Autocomplete, June 2026) is that the worth-it frame is universal. In orthodontics, patients search "are braces worth it," "is invisalign worth it," "are braces worth it as an adult," and "do i need braces or invisalign." In plastic surgery, the same shape appears at scale: "is a tummy tuck worth it," "is rhinoplasty worth it," "is a facelift worth it," "is botox worth it in your 30s," "is morpheus8 worth it." Whatever the procedure, the patient's first real question is the same one, and it is a question of certainty, not price or proximity.

The second pattern is just as telling: the constant "reddit" suffix. "Are dental implants worth it reddit," "is invisalign worth it reddit," "is a tummy tuck worth it reddit," "how to choose a plastic surgeon reddit" all show up verbatim in the data. Patients are deliberately routing around marketing to find a real person who has been through it. This is a credibility signal a practice cannot fake, and it explains why a polished homepage that affirms the obvious answer wins nothing. The patient already distrusts the affirmation. What earns the citation, and the visit, is content that does what the patient wishes the marketing would: state the honest tradeoffs, the durability, the candidacy, the cost rationale, in specific and verifiable terms. That is what an AI can quote, and what a wary patient can trust.

Why This Happens Before They Ever Call You

The order matters. By the time a patient searches "best cosmetic dentist near me" or "how to choose an orthodontist," they have already worked through the worth-it and comparison questions, often days earlier, on AI. The framework they will judge you against, what counts as a good provider, what the treatment should cost, what the alternatives are, was built during that earlier phase. If you were absent from it, you are now being evaluated against criteria a competitor or a forum thread helped set.

And the click that used to deliver the patient to you is disappearing. Pew Research found that when a Google AI summary appears, users click a result just 8% of the time versus 15% without one, and click a link inside the summary itself only 1% of the time (Pew Research Center, 2025). So even a top-ranking page can produce no visit. The patient reads the answer and moves on, carrying an impression of who the credible providers are, whether or not anyone earned a click. A practice that only invests in ranking is optimizing for a doorway that, on a growing share of these searches, the patient never walks through. The opportunity is to be inside the answer, named and trusted, not in the list beneath it.

Answer These Questions on Your Surface, or AI Answers Them With Someone Else's Content

An AI does not invent its answer to "are veneers worth it" out of nothing. It assembles it from the most specific, coherent, citable content available on the public web for that exact question. Right now, for most practices, that content does not come from them. It comes from a forum thread, a national directory, or a competitor who happened to publish a real answer. The patient's question gets answered either way. The only variable a practice controls is whether the answer carries their name and their framing or someone else's.

Closing that gap is not a volume problem, and it is not solved by posting more reels. It is a structural one: building, across the patient's real question clusters, the specific answers an AI can quote and a wary patient can trust. That means worth-it content with honest longevity and candidacy detail, true comparison content for the "this versus that" searches, cost content that gives a rationale rather than just a number, and provider-selection content that names the criteria a good patient should use. The bespoke strategic architecture that maps a practice's whole demand-capture surface, then builds the cinematic and written assets to fill it, is exactly the work that decides whether you are the answer or the absence. Before any of it is built, a diagnosis: which questions is your practice already absent from, and who is being named in your place. That is the gap a Vitals Audit is built to find.

The diagnostic frame

The patient asking AI "is this worth it" is doing the most important part of their decision before you know they exist. The practice that answered that question, on its own surface, in specific and trustworthy terms, is the one carried into the next search as a credible name. The practice that left the question to a forum or a competitor is evaluated against a framework it never helped set. The questions on this page come from our first-party Signal Mining, the same intelligence work behind our 834-post corpus of practice content. A Vitals Audit maps exactly which patient questions you are absent from, and who AI is naming instead, against three local competitors. Diagnosis before prescription.