Getting found for the specific procedure, not just "dentist near me"

Most practices chase the generic term anyway, because it has the biggest number next to it in whatever keyword tool the agency showed them. Volume is not demand. A term that returns every practice within eight miles, plus three ads and a map pack, is a term where you are the fourth-best option to somebody who has not decided anything yet.

The procedure searcher is a different person entirely. They know what they want, they are usually further along, and the case is usually bigger.

Why is the generic term the worst one to chase?

Because it is decided almost entirely by proximity and profile strength, and you cannot out-write proximity.

The ranking evidence is unusually clear on this. In Whitespark's 2026 Local Search Ranking Factors report, Primary GBP Category ranks first at 227 points and Proximity of Address to the Point of Search ranks second at 225. Physical Address in City of Search is fourth at 213. The two on-page factors most agencies sell against a generic term, keywords in the landing page title tag and keywords in the landing page headings, rank seventeenth at 146 and twentieth at 135.

That report is a survey of 47 local search experts scoring 187 factors, so it is opinion rather than test data, and Whitespark sells local SEO software. Grade it that way. But the ordering is not controversial and it is consistent with the large-n work: Darren Shaw's analysis of 1.8 million Google Business Profiles across 4,209 categories, published on Search Engine Land in August 2026, found the exact-match category outperforms every adjacent category, and that the pattern repeats for almost every category in the dataset.

Meanwhile the intent is thin. That searcher might need a cleaning, might be comparing, might be looking for their old dentist's new number. A booked appointment from that term is worth whatever a general appointment is worth.

A person searching "dental implant consultation" and a person searching "dentist near me" are not the same lead with different words. They are different economics.

Procedure searches also have shallower competition. The corporate group's location pages talk about "dental services." A practice with a real implant page, written by someone who places implants, has something to compete with that a generic page does not.

How do I find the procedure searches people actually type?

Start with your own data, not a keyword tool.

Pull your production report by procedure for the last twelve months. Sort by total production, not by count. The procedures at the top are the ones worth being found for, because they are the ones that pay for the effort.

Then listen to the phone. The words patients use are rarely the clinical words. They say "cap" for crown, "clear braces" for aligners, "gum surgery," "tooth pulled." Your pages should contain both the words you use and the words they use, because they are searching in their vocabulary.

Then check the search results themselves for each term. Type it on a phone in your city. Look at what Google returns: a map pack, a list of practices, an article, a health site. If the top results are national health information pages, the term is informational and it will not book anybody. Skip it.

There is now a fast test for that, and it is one of the more useful findings in local search. Ahrefs analyzed 146,122,391 desktop search results pages and found AI Overviews on 20.5% of them, but the distribution is what matters: 99.9% of the keywords that trigger an AI Overview are informational, against 4.3% commercial and 2.1% transactional. Health sits at 43.0% and medical YMYL at 44.1%, among the highest of any category. And only 7.9% of local searches trigger an AI Overview at all.

Read that as a sorting rule. "How much do dental implants cost" is informational, health-adjacent, and very likely to return an AI answer above the results, which means a practice page there is competing for a shrinking click. "Dental implants Alameda" is local and transactional and still returns a map pack and practice websites. The first is content you write to be quoted. The second is content you write to be called.

One caveat travels with those numbers. A 2026 Whitespark study of 540 manually run queries reportedly found AI Overviews on a far higher share of local business queries, because it used long conversational phrasings that almost always trigger them, while Ahrefs classified local by intent tag across a database dominated by short "[service] [city]" patterns that almost never do. Both are right about their own sample.

Also look for the modifiers people attach: cost, near me, same day, financing, "does insurance cover." Those tell you what the page has to address, and most of them are money questions rather than clinical ones. Answer the money ones inside your own limits: you may name the carriers you are in network with and offer to verify benefits. You may not tell anyone what their plan covers.

Which procedures deserve their own page, and what is one worth?

The ones with production behind them and a real answer to give. Usually four to eight, not twenty.

A page earns its place when it can answer the questions a person asks before booking that specific procedure: what it involves, how long it takes, what recovery looks like, what determines the cost, and how payment works. If you cannot fill a page without padding, that procedure does not need one yet.

Then price the page against the market, because the paid market is the only honest benchmark for what an inquiry is worth. LocaliQ's 2026 search advertising benchmarks put dentistry at a $72.97 average cost per lead and an $8.00 average cost per click, third most expensive of any category. Those are medians across thousands of US campaigns, published by a company that sells advertising management, and the 2026 sample size is not disclosed. Use it as a price signal, not a traffic model.

Here is the comparison, with your own numbers.

Count what each page produced last quarter. Implant page: 27 inquiries in 90 days, so 9 a month. Generic "dentist in [city]" landing page: 42 inquiries, so 14 a month.

Price them at the market rate. At $72.97 a lead, the implant page avoided $657 a month of media and the generic page avoided $1,022. On that measure the generic page wins.

Now weight by what they became. Your practice management software says a booked implant case produces $3,000 and a general new patient produces $340 in first visit production.

Multiply by what actually converted. 4 of the 27 implant inquiries became cases, so $12,000. 11 of the 42 generic inquiries became new patients, so $3,740.

That is the argument in one line. The page with fewer inquiries produced more than three times the money, and it is the page that is easier to rank because fewer practices bothered to write it.

Money is the part most practices skip and it is the part patients came for. You do not have to publish a fixed fee. You do have to explain what makes the number move, and what the payment paths are, including how membership and package plans get presented online for people without benefits.

The insurance question shows up on every procedure page too. Handle it the same way everywhere, and handle it carefully, because what a website can safely say about insurance is narrower than most practices assume. Name your participation. Never describe what a plan pays.

What makes a procedure page rank rather than just exist?

Being specific, being local, and being backed by a business Google can see is real.

Specific means the page is about one procedure and says so in the title, the first sentence, and the headings. A page trying to cover implants, bridges, and dentures ranks for none of them.

Local means the page names where you are and where your patients come from, in normal sentences. Geographic Keyword Relevance of Content ranks eighteenth at 142 points in the same Whitespark survey, which puts it well behind category and proximity and well ahead of nothing. It is worth doing properly and it is not worth a footer stuffed with thirty city names.

Real means the rest of your presence supports it. Your Google Business Profile, your hours, your address, your review flow. Procedure rankings sit on top of local ranking generally, which is why a practice that moves locations can lose ground it spent years earning even though the pages did not change.

And here is the tactic to refuse, because it is the first one anybody suggests for a procedure page. Do not coach patients to mention the procedure in their reviews. Sterling Sky ran a controlled test on it: a Christmas tree farm with no SEO work and a poor website, tested in summer when it would not otherwise receive reviews, had six people leave reviews containing target keywords against a base of 18 total reviews. Joy Hawkins's finding, verbatim, is that adding keywords to reviews did not help the business's local pack ranking on Maps, and that in the first instance rankings got worse. A second independent test using an invented word found the same.

The confusion this creates is worth resolving because agencies use it as a defense. Google does sometimes surface a "justification" snippet quoting a review that matches the query. Hawkins explains the ordering: those get pulled in when Google finds a matching review, but you have to rank first, and then the review text gets displayed. It is a display feature, not a ranking input.

Three reasons never to do it: it does not work, Google's own policy prohibits requesting that specific content be included in a review, and it makes your practice look strange to the patient you asked.

Reviews still matter enormously here, and dental has a constraint other industries do not. You can ask everyone, but you cannot answer publicly in a way that confirms someone was a patient, so reviews for dental practices sit on a specific HIPAA line that is worth understanding before you build the request flow.

What to do this week

Pull the twelve-month production report and write down your top five procedures by revenue.

For each one, check whether a page exists that is genuinely about that procedure alone. Most practices find two of five.

Search each term on a phone, in your city, and write down who ranks and whether an AI Overview appeared. If it is three national health sites under an AI answer, cross it off. If it is local practices in a map pack, you have a target.

Pick one procedure and write that page properly this month. One good page beats five thin ones, and the thin ones are actively working against you.

Then make sure the phone can handle what the page produces. If an automated system is answering, test it against real procedure questions before you send traffic at it, because most AI voice agents that work in a demo fail in the first week live.

Which procedure should you go after first?

Email me at eric@seod.com with your city and the three procedures you most want more of. I will search all three the way a patient would, look at who is ranking and how strong they actually are, and tell you which one is realistically winnable this year and which one is already locked up by somebody with a ten-year head start.

That is a real answer, not a pitch, and sometimes it is "none of these, your problem is the profile."

There is more on dental and med spa marketing if you want to keep going.