You can't buy the recommendation. Google says buying ads does not influence its organic AI answers, OpenAI says ads do not influence ChatGPT's answers, and ChatGPT's ads test keeps ads away from health topics entirely. When a patient asks an AI assistant which injector or surgeon to see, the names in the answer are earned. Keep the ads that book consults today, and put your next marketing dollar into being one of those names.
What do Google and OpenAI actually say about ads and AI answers?
Both companies have put it in writing. Google's ads help page asks whether paid ads affect what its AI says in organic responses and answers no: buying ads doesn't influence organic AI answers, and being cited organically doesn't affect your ad auction eligibility. Ads in AI Overviews and AI Mode can sit within or alongside the answer, but they are labeled Sponsored and visually set apart from the organic content.
OpenAI's post announcing its ads test, first published February 9, 2026, says ads do not influence the answers ChatGPT gives and that answers are optimized for what is most helpful to the user. Ads are labeled sponsored and visually separated from the answer. The same post says that during the test, ads are not eligible to appear near sensitive or regulated topics like health, mental health or politics, and are only shown to logged-in adults on the Free and Go plans.
Read those two statements together and the conclusion for a practice is plain. On Google, you can rent the space next to the answer. In ChatGPT, for most patient questions, there is no space to rent. Either way, the shortlist a patient reads is built from what the engine finds about you, not from your budget.
Why the answer matters more than the slot next to it
Because fewer patients scroll past it. Pew Research Center tracked the browsing of 900 US adults in March 2025 and found that Google users who saw an AI summary clicked a traditional result in 8% of visits, against 15% when no summary appeared. They clicked a link inside the summary in just 1% of visits, and ended their session on 26% of pages with a summary versus 16% without.
That is general search, not medical search, and Pew measured clicks, not consults. But the direction is clear: when an answer appears, more of the decision happens inside it. A patient who reads three named practices with a sentence on each may never see your landing page at all.
Patients are not reading those answers blindly either. In Pew's August 2025 survey of 5,153 adults, 53% of people who had seen AI summaries had at least some trust in them, and only 6% trusted them a lot. That is the honest picture: patients treat the answer as a starting shortlist, then check it. The check is your reviews, your provider bios and your directory profiles, which is the same material the engine read to build the shortlist.
And they are asking health questions in these tools now. KFF's poll of 1,343 adults, fielded February 24 to March 2, 2026, found 32% had used AI chatbots for health information in the past year. Among those users, 41% said a reason was to look things up before deciding whether to see a provider. KFF did not ask how patients pick a specific practice, so treat that as a signal of where research starts, not a share of your consults.
What earns a practice a place in the answer?
The same things a careful patient would check. Google says there are no additional requirements to appear in AI Overviews or AI Mode: a page has to be indexed and eligible to show with a snippet. It also says these features may use "query fan-out," running several related searches across subtopics and sources to build one answer. A question about lip filler near you can pull in your procedure page, your Google Business Profile reviews, a directory listing for your injector and a forum thread, all at once.
So the work is not one channel. It is making the facts about your practice consistent and easy to confirm wherever an engine looks:
- Provider pages that state each surgeon's or injector's credentials, procedures and locations in plain text.
- Procedure pages that answer the questions patients ask your coordinators, including candidacy, recovery and how consults work.
- Directory and review profiles where names, credentials and services match your site.
- Reviews that describe the actual experience, gathered within the FTC's rules on fake and incentivized reviews.
None of that can be bought on an auction. All of it compounds: a corrected credential on one directory keeps working every month without a new bid.
A worked example
For example, suppose a two-location med spa spends an illustrative $6,000 a month on Google Ads and gets most of its booked consults from those campaigns. The owner reads that ads now appear in AI Overviews and asks the agency to expand into those placements, expecting to "show up in AI."
The campaigns become eligible, the ads start appearing as Sponsored units beside some answers, and spend rises. But when the practice manager runs ten patient questions in Google, ChatGPT and Perplexity, the answers still name two competitors and a chain. In ChatGPT, under OpenAI's stated policy, health questions show no ad at all. The med spa is paying for space next to a recommendation that goes to someone else.
Now suppose the same owner leaves the working campaigns alone and spends the next quarter on the off-site facts: the lead injector's credentials made consistent across directories, a procedure page for each top treatment, and a steady review request after every visit. The ad results stay where they were. Whether the practice starts appearing in answers is something the manager can see directly by running the same ten questions again. This is an illustration of the decision, not a client result, and no one can promise the second path gets the practice named.
But ads still book consults, and AI answers can't be measured
This is the strongest objection, and the first half is true. Paid search is measurable to the booked consult. You can set a cost per consult, watch it weekly and shut off what fails. AI answers give you no impression count in ChatGPT, no bid and no guaranteed spot. If your ads book consults at a cost you can live with, cutting them to fund something slower would be a mistake, and I would not advise it.
The second half does not hold. AI answers can be measured, just not by an ad platform. Run a fixed set of patient questions on each engine several times a month and record whether your practice and each provider is named, how they are described and which sources are cited. That gives you a share-of-answer number you can track against last quarter. It is rougher than cost per click. It is far better than not knowing.
And the objection misreads the trade. The argument is not ads or answers. It is where the next dollar goes. Google's own help page says the ad and the organic answer are separate systems. Extra ad spend buys more ad. It does not buy a mention in the answer next to it, and in ChatGPT's health conversations it buys nothing, because OpenAI does not show ads there. If you want to be named, the only lever is the one that feeds the answer.
One caution for practices that do expand into AI ad placements: Google says some placements include an AI explainer written by Gemini, and advertisers can't write, edit, approve or opt out of it. If your state board or the FTC limits the claims you can make, read those explainers before you scale.
How to check where you stand
Pick ten questions a patient would ask about your top three procedures, with your city in them. Run each one three times in ChatGPT, Google and Perplexity. Count how often your practice and each provider is named, and note whether a Sponsored ad appeared and whose it was. If competitors are named and you are not, more ad spend will not close that gap. Our guide to what AEO means for a medical practice explains what to fix first, and our take on replacing your SEO agency covers who should do the work.
If you would rather have it run for you, our free AI visibility audit tests your real procedure and provider questions across the major engines and shows which sources the answers lean on. More on how we work with practices is on the medical practices page.
Frequently asked questions
Can I advertise my practice inside ChatGPT?
Do Google ads in AI Overviews hurt my organic chances of being cited?
Who writes the text Google shows next to my ad in AI results?
How long does it take to start being named in AI answers?
Sources
- Testing ads in ChatGPT · OpenAI
- About ads in AI-powered experiences in Google Search · Google Ads Help
- AI features and your website · Google Search Central
- Google users are less likely to click on links when an AI summary appears in the results · Pew Research Center
- Americans have mixed feelings about AI summaries in search results · Pew Research Center
- Poll: 1 in 3 Adults Are Turning to AI Chatbots for Health Information, Equaling the Share Who Use Social Media for Health · KFF