Split the budget at the county line before you split it by procedure
Metro Nashville and Davidson County run as a consolidated government while the surrounding counties do not, and the buyer changes at the same line the government does.
A single campaign covering the metro will hand its budget to whichever geography clicks cheapest that week. That is almost never the geography with the best patients, because cheap clicks tend to come from the places where intent is thinnest.
Give each county its own campaign and its own budget so that neither can borrow from the other. You then get to see, without arithmetic, what a Williamson County patient costs against a Rutherford County patient, and you can move money on purpose.
Set location targeting to people present in the area rather than people present or interested in it. The interest setting will show your ads to someone in another state reading about Broadway, and dental demand does not travel that way.
Name the places you will not serve and exclude them on day one. If nobody is driving from Mount Juliet to your Bellevue office, stop paying for that click before it becomes a habit.
Performance Max will spend outside every line you drew
Automated campaign types are built to find the cheapest conversions available, and in a dental account the cheapest conversion is usually somebody who was going to call you anyway.
The pattern is consistent. Automated inventory absorbs branded searches, display placements and low quality traffic, then reports a flattering cost per conversion built largely on demand you already had.
Start with search campaigns you can read: exact and phrase match, tight geography, negatives you maintain weekly. When the account is producing reliable booked-patient data, and only then, consider automated inventory with brand exclusions applied and a separate budget you are willing to lose.
Pacing matters as much as structure. A dental practice has finite chairs and finite hygiene hours, so an account that outruns the schedule is buying calls the front desk cannot place. Set the daily budget against the appointments you can actually seat, then raise it when the schedule shows room.
Check the search terms report yourself, or make sure whoever runs the account can show it to you. It is the one place where a tidy dashboard and reality can be compared.
Buy the no-insurance searches most dental accounts skip
The tourism and short term rental economy downtown supports a large service workforce, and a meaningful share of it is paying out of pocket and searching in price language.
Queries about cost without insurance, payment plans and membership plans look low value in a keyword tool, so many accounts never bid on them. The patients behind them are real, they are usually unattached to any practice, and they convert when the answer is straightforward.
Bidding on them only works if there is somewhere to send them. A page that explains what a new patient exam includes, what a membership plan covers and how payment works will hold that click. A generic homepage will not.
The front desk has to be able to answer the same question in the same words. If the ad and page say one thing and the phone says call us for details, you paid for the click twice and got nothing.
Keep this as its own campaign with its own budget. It answers a different question from the implant campaign and it should never be compared to it on cost per click.
An ice storm empties Thursday and search is how you refill it
Real winters here mean occasional closure days and cancellation clusters, and paid search is the fastest lever available for putting people back in an empty schedule.
Ice is an ordinary consideration in Middle Tennessee, not a rare event. A morning of it will cancel a day, and the practices that recover fastest are the ones that treat the empty chairs as an inventory problem the same afternoon.
Hold a small reserve inside the monthly budget rather than spending flat. When a day empties, raise bids on the immediate-need terms for a few days and pair it with texts to the recall list. When the schedule fills, put the budget back.
Weather also moves emergency demand. People slip on frozen steps and break things, and searches for urgent dental care rise with it. Whether that traffic is worth buying depends on whether you can see anyone the same day, which is a scheduling decision before it is a bidding one.
The same reserve logic covers the rest of the dental calendar: benefit maximums running out late in the year, orthodontic consults clustering in summer, families booking around the school calendar.
Growth corridors and settled neighborhoods reward opposite bidding
In Murfreesboro or Mount Juliet you are usually reaching somebody who has no dentist yet, and in Belle Meade or Green Hills you are asking somebody to leave one.
Those are different sales, so they need different ads. On the growing edge, the message that works is availability: accepting new patients, appointments this week, what a first visit involves. There is no incumbent to displace, only an absence to fill.
In the settled neighborhoods, the searcher already has a dentist and is looking for a specific reason to move, usually a procedure, a schedule problem or a bad experience. Ad copy there has to be about the thing they want, not about being open.
Your tolerance for cost per click should differ too. Displacing an existing relationship generally takes more clicks per booking than catching somebody with no relationship at all, which is an argument for spending more per click in one place and more total in the other.
Do not average the two into one target and then wonder why performance looks mediocre everywhere. Report them separately or you will make decisions on a number that describes nothing.
Judge the account on production per new patient, not on cost per lead
A form fill is not a patient, a booked appointment is not a kept one, and only the last of those pays for anything.
Track the chain end to end: click, call or form, booked appointment, kept appointment, treatment accepted. Every stage loses people, and the stage where you are losing them determines whether the fix is in the account or at the front desk.
Feed the booked appointment back into the platform as the conversion, and where you can, feed value with it by procedure category. An implant consult and a hygiene appointment are not worth the same, and an account optimizing to undifferentiated conversions will drift toward whatever is cheapest to generate.
Once that data exists, the county split starts paying for itself. You can see that a booked patient from one county costs more and produces more, or costs less and cancels more, and neither of those shows up in a metro average.
Nobody can promise you a cost per booked patient before the account has run. What is reasonable to expect is that the number is measured honestly, reported by geography, and used to move budget rather than to decorate a report.
Questions we actually get
- How much should a dental practice in Nashville budget for paid search?
- It depends on how many chairs you can fill, which counties you want to reach and what you are selling. The honest starting point is capacity, not a benchmark: work out how many new patients per month you can seat, then build a budget aimed at that number and adjust once real booking data exists.
- Should we run Google Ads and Local Services Ads together?
- They buy different things and can run alongside each other. What matters is that they are measured separately, because mixing them produces a blended cost that hides which one is actually filling chairs. Eligibility and requirements for Local Services vary and are worth confirming directly with Google before budgeting for it.
- Our agency reports a great cost per lead. Why are we not busier?
- Usually because a lead is being counted at the wrong point. Form fills, chat sessions and short calls all get counted as conversions, and none of them are a patient in a chair. Ask for the same report by booked appointment and by kept appointment, split by county, and the picture normally changes.
- Do we need separate campaigns for Franklin and East Nashville?
- If you draw meaningful patient volume from both, yes. They are different counties, different buyers and different competitive sets, and one campaign covering both will spend where clicks are cheapest rather than where patients are best. Separate budgets also let you turn one down without touching the other.
- Is it worth bidding on our own practice name?
- Often yes, because competitors can appear above your listing on your own name and the clicks are usually inexpensive. It is worth measuring rather than assuming. Run it for a period, look at what happens to total branded traffic, and keep it only if the numbers support it.
