Fort Worth and Uptown should never share an ad group
An hour of freeway separates the two anchor cities, and no single bid is correct for both.
Dallas and Fort Worth are distinct markets with distinct downtowns, competitor sets and prices. A campaign covering both averages two markets into one number and then optimizes toward the average, which serves neither.
Radius targeting compounds the error. A circle drawn around a Southlake or Grapevine office crosses Tarrant, Denton and Dallas counties before it finishes. Target by city and ZIP instead, chosen from where your current patients actually live.
Pull those addresses out of the practice software. You will usually find the real service area is smaller and lumpier than the map you have been buying, with a long tail of one-off patients you should not pay to repeat.
Set location targeting to people present in your chosen areas rather than people showing interest in them. Otherwise you buy clicks from someone in another state reading about dentists in Dallas.
Then name the exclusions out loud. Deciding you will not serve Denton County is a budget decision, and making it on purpose is cheaper than discovering it in the search terms report.
Bid on the procedures your chairs are open for this month
Account structure should follow open capacity, not the services list on your website.
Every practice has a services page and a schedule, and only the schedule tells you what to buy. Paying for hygiene clicks while the hygiene column is booked six weeks out is spending money to lengthen a wait.
Build campaigns around the two or three treatments you want more of, each with its own budget, its own landing page and its own definition of a good outcome. Implants, adult orthodontics, an associate's new day, whatever the gap is.
Review it monthly with whoever runs the schedule. Capacity moves when a provider adds a day or a hygienist leaves, and the account should move with it.
Emergency terms deserve their own treatment. They convert fast, they are expensive, and they are worthless if the phone rings out at lunch. Run them only during hours somebody answers.
Exclude the students, the job seekers and the aligner-kit shoppers on day one
Dental keywords attract three audiences who will never sit in your chair.
Job and training searches share almost every word with your service terms. Dental assistant jobs, hygienist salary, programs at area colleges, continuing education. None of them book.
Direct-to-consumer aligner brands, whitening kits and home remedy searches pull the same terms from the other direction. Someone comparing mail-order kits is not shopping for your chair.
Free and reduced cost care is the third group. Dental school clinics, charity days, sliding scale searches. Those people have a real need and you are not the answer to it.
Write the negative list before launch, then read the actual search terms report weekly for the first month and add to it. Broad match keeps finding new ways to spend, and the report is the only place you see it.
Some phrases are worth keeping despite looking wasteful. Someone searching what an implant costs is early but serious, and that click is usually cheaper than the one they make three weeks later.
Judge the account in chair hours filled, not leads received
A form fill is not a purchase, and two booked appointments can be worth very different amounts.
Standard reporting stops at conversions, so cost per lead flatters an account producing cheap and useless inquiries. Twelve forms from people wanting free care beats four booked exams on that scoreboard, and loses on every real one.
Track four numbers instead: calls answered, appointments booked, appointments kept, and what those visits produced. The gap between the first and second is a front desk problem. The gap between the second and third is a reminder problem. Both cost more than your bids.
Feeding booked appointments back into the platform as offline conversions teaches the bidding what a good click looks like, which matters more as more of the auction runs on automation.
Handle patient data carefully when you do it. Keep clinical detail and identifiers out of advertising platforms, use the anonymized paths your vendors provide, and treat the whole question as one worth raising with your own counsel before you build it.
Chair time is the honest unit. An account that fills the surgical column and starves the hygiene column may look worse per lead and be worth far more.
December expiry and January network changes are two different budgets
Dental demand runs on a calendar, and flat monthly pacing ignores both ends of it.
Late in the year, patients with remaining benefits and a met deductible finish treatment they have been putting off. Messaging shifts toward completing planned work, and the constraint is chair time, not demand.
January is a different buyer. Plans changed, networks changed, and households across Collin and Denton counties who arrived with a new employer are finding out where they can go. Insurance and new patient terms carry the weight.
Summer moves pediatric and orthodontic volume around school schedules, and the family searching in July is booking for August.
Set the pacing plan in advance rather than reacting to a slow week. Reactive budget changes inside an automated bidding system generally cost more than they save.
None of that means guaranteed seasonality for your practice. Look at your own production by month first, then pace to what your own book actually does.
Three funded campaigns beat eleven that never learn
One practice budget split eleven ways produces noise and a monthly report nobody can read.
The temptation is a campaign for every procedure crossed with every suburb. It feels precise and it starves everything, because none of the segments accumulate enough conversions to be judged or to teach an algorithm anything.
Start narrow. One campaign on your core new patient terms in your real service area, one on the treatment you most need to fill, one brand campaign to hold your own name. Add a fourth when one of the three has earned it.
Search first, and be honest about what other channels do. Paid social introduces elective treatment to people who were not looking. It rarely fills a hygiene column, and judging it by the same cost per booking is how practices talk themselves out of it too early or into it too heavily.
Give any change enough weeks to mean something. A single practice in one metro generates modest weekly volume, and the temptation to declare a winner after five days is the most expensive habit in the account.
Questions we actually get
- What should we budget?
- We will not quote a figure before seeing your market and your capacity. The useful way to set a floor is backward: decide how many new patients a month you have room for, look at what your practice earns from a new patient in the first year, and work out what you could pay to acquire one and still be glad. Then check whether the auction in your part of the metro fits inside that. Sometimes it does not, and paid search is the wrong first move.
- Should we bid on our own practice name?
- Usually yes, and the reason is defensive. Competitors and directories bid on practice names, and brand clicks are typically among the cheapest in the account. If nobody is bidding on you and you own the organic result outright, it is a fair thing to test turning off.
- Do Local Services Ads apply to dental?
- Availability by category and location changes, so it is worth checking current eligibility for your specialty rather than assuming. Where a verified format is available it generally changes what the top of the page looks like, and we would test it as its own channel rather than folding it into search reporting.
- Why are you asking for access to our scheduling data?
- Because cost per lead is the wrong number and we cannot compute the right one without knowing what got booked and kept. It does not require clinical records. Appointment counts, treatment category and kept status are enough, and we will scope exactly which fields are shared before anything is connected.
- Can you guarantee a cost per new patient?
- No, and anyone who does is quoting you a number they cannot control. Auction prices, competitor budgets and your own answer rate all move. What we can do is report the real number every month and change the account when it drifts.
