Two rejected ads in week two can cost an agency a dental or real estate client faster than a bad month of targeting. Should that agency hire a vertical specialist who knows the rules by heart, or a generalist who ships more creative? Smart operators disagree because Meta publishes the restrictions for everyone, yet a missed restriction can still wreck a launch.

I have sat in both chairs. Written rules weaken the specialist’s claim to secret knowledge; costly mistakes weaken the generalist’s claim that a checklist is trivial. I would hire the generalist with a real creative engine and a written compliance process for most small agencies. One dental client profile changes my answer.

The rules neither side gets to ignore

Dental is not a housing-style Special Ad Category. Meta’s four categories are housing, employment, financial products and services, and social issues, elections and politics. A dental ad promoting financing can raise a separate category issue (the four categories are listed here). The rule dental advertisers meet every week is the personal-attributes restriction: an ad must not assert or imply someone’s physical or mental health condition. New implant consults available in Round Rock describes an offer. Embarrassed by your missing teeth? presumes something about the reader.

Cosmetic-procedure ads bring another check: before-and-after material requires 18-plus targeting and cannot use language that insults someone’s appearance (the age and language restrictions). The less visible dental risk sits on the website. Meta will not sign a Business Associate Agreement for its ad platform, so protected health information from an appointment or intake page must not reach the Pixel (why medical and dental practices need to examine Pixel placement). Review the site as well as the ad.

Real estate is simpler and stricter. A listing ad belongs in the Housing Special Ad Category, and that declaration changes the available targeting: age runs from 18 to 65-plus, gender is All, ZIP code targeting and exclusions are unavailable, lookalikes are unavailable, detailed targeting is limited, and city or pin radiuses expand to roughly 15 miles minimum (Housing targeting restrictions).

Copy has another layer. Fair Housing concerns extend beyond what automated Meta review catches. Describe the house, lot, price and location, not the buyer you picture living there (the property-first copy rule). Perfect for young families can clear a review and still create a problem. I used to tell clients targeting would fix real estate. I was wrong. Under Housing rules, you cannot target your way out.

Position A: Pay the specialist to avoid week-one mistakes

The specialist has already paid the tuition you are about to pay. A dental team has seen the question that implies a condition, the before-and-after ad with the wrong age gate, and the financing line that needs another compliance check. It writes Implant consults open this month in Round Rock on the first draft instead of learning on your client’s budget.

It also knows to inspect what happens after the click. A Pixel firing on an intake page can send health information to a platform that will not sign the agreement needed for that transfer. A team that runs dental accounts every week knows to look for that setup during launch QA. A generalist absorbed in ad production may not ask until the client asks why the retargeting feels creepy.

The specialist’s stronger argument, though, is commercial. Dentists do not buy CTR; they want started cases. Dental lead economics can be punishing: one cited benchmark puts the leads objective at $61.56 per lead, $5.70 CPC and 1.62% CTR. A specialist knows why an implant campaign cannot be judged by the same cheap-lead target as a cleaning offer, and why cost per started case matters more than a tidy dashboard.

Real estate flips the problem. A cited benchmark puts leads at $13.74 against a $27.39 cross-industry average. Cheap valuation-form leads may still be months from a listing. The specialist follows the lead beyond the form instead of congratulating itself on CPL.

That experience informs the less glamorous playbook: one dental procedure per budget until it works, 4 to 8 creative angles for that procedure, an 8-to-15-mile draw for general work or 20 to 40 miles for full-arch, and an instant form backed by a call in under five minutes. Those choices concern the offer, the catchment area and the handoff, not some secret audience toggle. You are buying fewer week-one surprises, not targeting secrets.

Position B: Hire the generalist that keeps making new ads

The generalist’s best argument begins with an uncomfortable fact for specialists: the rules are published. Meta gives examples for personal attributes. The Housing declaration changes targeting in Ads Manager. Meanwhile, the fine-grained audience work that once made a niche buyer look indispensable carries less weight. Advantage+ Audience is the default for most objectives, and Meta can expand beyond suggested detailed interests.

I watched careful interest stacks stop mattering. When the platform gives buyers fewer fine-grained levers, the useful work shifts toward what they can still vary: the offer, the words, the image and the follow-up. A generalist that produces distinct ads quickly has more chances to find the message that gets a qualified response.

Housing makes the case especially clear. With age constrained, ZIPs unavailable and radius minimums near 15 miles, the creative has to do more of the sorting. Show the home, state the price and name the neighborhood in the copy and walkthrough. Those details help a relevant buyer recognise the listing without trying to define who is allowed to see it.

Dental benefits from the same discipline. A 4-to-8-concept baseline for one procedure gives an account real alternatives to test. A generalist shop that ships that volume every two weeks across several kinds of accounts may learn new hooks faster than a specialist reusing last year’s winner for five dentists in different cities. Specialists sell memory. Generalists sell testing speed. In a restricted account, fresh, compliant creative is a lever you can still pull.

The rebuttal: where each pitch breaks

The specialist pitch breaks when experience turns into repetition. Some shops marketed as specialists serve several local lead-generation niches under one roof, from roofing and HVAC to dental, med-spa, real estate and legal. There is nothing wrong with that on its own. The problem comes when Client A’s winning hook becomes the template for Clients B through Z. Your dentist and the practice two towns over run near-identical smile ads. The report celebrates steady CPL while cost per started case drifts.

I have audited accounts like that. The compliance was clean; the ads were tired. Knowing what passed review last year is not a creative strategy for this month. Ask to see dated output, not just a list of verticals.

The generalist pitch breaks at the details that can stop a launch. A team can ship fifty angles a week and still miss a financing claim, publish a listing without the Housing declaration, or put the Pixel on an intake page. More creative only means more chances to make the same mistake if nobody owns the policy check.

Then there is the handoff neither side likes to own. The dental playbook calls for instant forms followed by a call in under five minutes and a seven-day cadence. A provider can report a fine CPL while nobody calls the lead promptly. The case never starts. Ask who owns the first seven days after the lead, not just the first seven days after launch.

Five questions to put to either provider

Send these before you sign. A good provider answers in specifics; a weak one answers in confidence. For more on the production side, read who should make your Meta ads alongside the questions.

  1. Can you rewrite these two lines? Give them Tired of hiding your smile? and Perfect for young families near St. Marys. Look for property- and offer-led alternatives, such as Implant consults open this month in Round Rock and 3-bed brick ranch, fenced lot, 12 minutes to downtown. If they spend the call defending the original wording, keep looking.
  2. Can you walk me through a Housing launch? Listen for Special Ad Category selection, age from 18 to 65-plus, gender set to All, no ZIPs or lookalikes, and a radius of at least 15 miles. A promise of tight ZIP targeting for a listing is not expertise.
  3. Where does the Pixel fire on a dental site? The answer should identify pages to exclude: booking, intake, portal and condition pages. It should not propose patient-list uploads for lookalikes or blanket retargeting of every site visitor.
  4. How many new angles does each account get every two weeks, and who makes them? Ask for 4 to 8 starting concepts for one procedure or listing type, then fresh variants. Request last month’s dated output from two accounts. This is where many white-label models for small agencies fall over: a fee without a creative rota.
  5. What number do you report, and who handles the lead? Dental needs cost per booked consult or started case; real estate needs cost per booked appraisal or listing appointment. Ask who makes the first call and who follows up for seven days. CPL alone cannot answer either question.

That conversation takes about twenty minutes. Run it before you put your agency name on the provider’s work.

My verdict: buy creative speed, with compliance in writing

For an agency with three dental clients and two realtors, I would choose the generalist with a genuine creative engine. Restrictions narrow the targeting options available to everyone, so a partner that keeps producing distinct, policy-safe ads has more to work with than one trading mainly on remembered settings. A rejected ad may cost two days of review and an awkward call. A fresh angle that earns better booked consults or listing appointments can keep working after that call is forgotten.

But the condition is non-negotiable: the compliance process must exist on paper before spend starts. I want a personal-attributes wording check, a before-and-after age-gating check, a financing-language flag, a Housing declaration checklist and Pixel placement rules. If the provider cannot show me those documents, I do not care how many concepts it shipped last Tuesday.

The same suspicion of paying for work nobody can show you is why I prefer a flat-fee model such as groas’s to percentage-of-spend theater. The fee structure is not a substitute for checking a Meta provider’s actual creative output and compliance process. Ask for both.

The one client I would give to a specialist

Pay the specialist for a full-arch or high-ticket implant practice where a single started case covers a month of fees. Say the client spends $20k a month trying to book six consults. A sloppy financing line or a Pixel on the wrong intake page threatens far more than a routine ad rejection. A dental-only team that works toward cost per started case, understands the 20-to-40-mile draw and runs the sub-five-minute call routine earns its premium there.

For the small agency with two general dentists and a realtor who mainly needs listing alerts, I would not pay that premium merely for reassurance. When one case carries the month, buy the specialist’s memory. Otherwise, buy testing speed and insist on the checklist.

I would rather explain a rejected ad I saw coming than explain an account that quietly tested nothing new. For most agencies, the answer is the generalist with a creative engine, held to a written compliance checklist. The high-ticket implant practice is the exception. If you are adding your first few Meta clients without hiring, this 60-day playbook for one manager running three clients offers a way to structure the work.