Dental marketing ROI by channel: SEO vs PPC vs referral

A practice putting $4,000 a month into Google Ads is buying roughly 296 new patients a year at $162 a head, and the morning the card gets declined, that pipeline goes to zero. That is not a marketing budget. That is rent, and the lease renews every 30 days.

Our thesis, and plenty of owners will fight it: for a single-location general practice under about $1.5 million in collections, paid search should be the last channel you fund, not the first. The standard agency recommendation (split the budget across search ads, SEO and "reputation management" so every box is checked) is how owners end up paying $162 a head forever for patients they could have owned outright for $83. The three channels are not three versions of the same thing. One of them you rent, two of them you buy, and the accounting treatment should not be identical.

$72.97 Buys a Lead, Not a Patient

Start with the only channel where the numbers are public and current. LocaliQ's 2026 search advertising benchmarks put Dentists & Dental Services at an $8.00 average cost per click, a 5.66% click-through rate, a 10.67% conversion rate, and $72.97 cost per lead.

That $72.97 is where most owners stop reading, and it is the most expensive place to stop. A "lead" in that dataset is a form fill or a tracked call. It is not a kept appointment. Between the lead and the chair sit three leaks: the lead who never answers the callback, the lead who books and no-shows, and the lead who shows up wanting a $59 cleaning special and nothing else.

Be generous and assume you convert 60% of leads to a booked appointment and keep 75% of those. That is a 45% lead-to-patient rate, which is better than most front desks run without a dedicated phone protocol. Your real acquisition cost is $162 per new patient. At $48,000 a year in ad spend you are buying about 658 leads and 296 patients.

Now the part the dashboard will not show you: that $162 does not improve with tenure. Year three costs the same as year one, adjusted upward for click inflation.

What a New Patient Is Actually Worth in Year One

Every channel comparison dies without a per-patient value, and most of the numbers floating around the industry are vendor marketing. So build it from the cost side instead.

BLS OEWS data for Offices of Dentists puts the median hourly wage for general dentists in that industry at $79.95 (May 2023), across 109,280 jobs, and the Occupational Outlook Handbook reports median annual pay of $176,110 as of May 2025. If doctor compensation runs 30% of collections, every doctor hour has to collect about $266 just to clear the comp line.

A typical new patient in year one consumes roughly 1.5 doctor hours, which carries about $400 of collections. Add the new-patient exam, radiographs and two hygiene visits at about $350, plus one restorative appointment at about $300, and you land near $1,050 in first-year collections. That is a modeled figure, not a survey result, and your own ledger will differ by specialty mix and fee schedule. Use it as a floor for the arithmetic below and replace it with your real number when you run this yourself.

At $1,050 in first-year collections, a $162 acquisition cost is 15.4% of year-one revenue from that patient. That is survivable. It is not good.

The 36-Month View: $48,000 a Year, Spent Three Ways

Here is the scenario that should decide your budget. One practice, $48,000 a year available for growth, three years, two strategies.

Strategy A: all paid search. $48,000 a year at $162 per kept new patient is 296 patients annually, 888 over three years. Total spend $144,000, blended cost per patient $162. On day one of month 37, if you pause the account, you get zero.

Strategy B: $2,500 a month to site and organic search, $1,000 a month to review and referral systems, $500 a month to paid search for brand defense and emergency terms only.

• Organic at $30,000 a year: call it 5 new patients a month in year one while pages age and the site gets fixed (60 patients, $500 each), 18 a month in year two (216 patients, $139 each), 22 a month in year three (264 patients, $114 each). Three-year total: 540 patients for $90,000, or $167 each.
• Reviews and referral at $12,000 a year (software, patient appreciation, staff time, a real ask script): 9 a month in year one, 12 in year two, 15 in year three. That is 432 patients for $36,000, or $83 each.
• Paid search at $6,000 a year: 37 patients annually, 111 over three years.

Strategy B delivers 1,083 new patients over the same 36 months on the same money. That is 195 more patients than the all-paid plan. At $1,050 in first-year collections, the gap is $204,750 in production.

Look at year one in isolation and Strategy A wins by a mile: 296 patients against 205. That is exactly why owners pick it, and exactly why they are still picking it in year five. The crossover happens in month 14 or so, and almost nobody holds the line that long.

Then look at year four. Strategy B can cut the organic retainer to $1,200 a month in maintenance mode and keep most of that 264-patient run rate, which puts cost per patient near $55. Strategy A's year four costs $48,000 again.

What Most Owners Get Wrong: Running SEO Through the Marketing Line

The specific mistake, and it is concentrated among owners five to twelve years in who have just added a second location or a second associate: they book organic search as a monthly marketing expense and judge it on a monthly report. That framing guarantees they cancel it.

A $2,500 retainer that produced 5 new patients last month looks like a $500-per-patient disaster next to a paid account producing at $162. So in month seven or eight, right before the compounding starts, the owner moves the organic money into ads where the dashboard looks better. We have watched that same decision get made under three different agency logos.

Organic search and a review engine are capital projects. You are building an asset that produces patients after you stop paying, which is the definition of equity, and it shows up in a valuation multiple in a way a paid account never will. A buyer pays for a practice that generates 20 new patients a month from its own name. A buyer discounts a practice that generates 20 new patients a month from a Google Ads account, because that cost transfers with the deal.

Our view: if you cannot commit 18 months to an organic program, do not start one. Put every dollar in ads, accept $162 forever, and stop pretending otherwise. The half-measure (nine months of SEO, then cancel) is the single most reliable way to waste $22,500.

Referrals Run Through Google Now

The old referral channel was a patient telling a neighbor. That channel still exists, but it has been routed through a review platform, and the routing changes how you invest in it.

BrightLocal's 2026 Local Consumer Review Survey found that 97% of consumers read online reviews for local businesses and 49% trust online reviews as much as a personal recommendation from someone they actually know. Read those two numbers together and the implication for a practice is concrete: a referred patient still checks your reviews before calling, and a stranger reading 180 reviews is behaving roughly like a referral.

That makes review volume the cheapest patient acquisition work available to a general practice, because the raw material is patients you already treated. The spend is operational, not media: a tablet at checkout, a text sequence that fires two hours after the appointment, and a hygienist who asks. At $1,000 a month all-in, including staff time, we modeled $83 per patient above. We would defend that number as conservative.

It also compounds in a way ads cannot. Every new patient from any channel is a potential review, so the paid and organic programs feed the referral engine. Run the ads without the review system and you are paying $162 for an asset you throw away.

Why 45% Is the Number That Should Set Your Mix

ADA Health Policy Institute figures put national dental expenditures at $189 billion in 2024, up 3.6% from 2023, while only 45% of the U.S. population had a dental visit in the previous 12 months.

That second number is the one that should set your channel mix. Paid search reaches people who are typing "dentist near me" today, which is a thin slice of that 45%. It does not reach the 55% who have not been to a dentist in a year and are not searching, because nothing has hurt yet. Those people get activated by someone they trust, in person or in a review they read while deciding whether to bother. The growing side of a $189 billion market is not the people already in the chair rotation. Ads cannot find the rest of them, and that is a structural ceiling, not a bidding problem.

When We'd Kill the Paid Account Outright

Set a hard rule: if cost per kept new patient exceeds 20% of first-year collections, the account closes. On $1,050 that threshold is $210. At the benchmark $8.00 CPC you are at $162 and you keep going.

In a dense metro where dental clicks run 50% above benchmark, cost per lead goes to about $109 and cost per patient to $243. That is 23% of year-one collections before a single crown gets seated, and we would not take that deal. We would cut to brand-term defense and emergency keywords only, hold the spend near $500 a month, and move the rest into the two channels that accumulate.

One exception, and it matters: a brand-new practice or a de novo second location has no name, no reviews and no aged pages. Buy patients at $162 or even $243 for the first 12 months. You are not buying ROI, you are buying the raw material for the review engine.

Three Numbers to Pull Before Friday

Pull your actual lead-to-kept-appointment rate from the last 90 days of tracked calls, not your agency's lead count. If it is under 45%, fix the phone before you touch the budget, because every channel improves at once. Then calculate your true cost per kept new patient by channel and test it against 20% of your real first-year collections per patient. Finally, count how many reviews you collected last month: if it is under 15, that is the highest-return hour of work in the building this week.

Marketing spend lives inside your overhead, and a $48,000 annual program is a percentage point or two of overhead on a million-dollar practice, which is the difference between a healthy practice and a tight one. If you want to see where that spend sits against the rest of your cost structure before you reallocate it, run your own numbers through the overhead calculator and check what your marketing line is doing to your margin.

What we are watching next: whether that 10.67% paid conversion rate holds as AI answer panels absorb more local search clicks, because if it drops to 7%, cost per patient goes to $248 and the kill rule fires in most markets. What would change our mind on the ordering: a practice showing review-driven acquisition cost climbing past paid search as platforms monetize local placement harder. If that happens, the rent starts looking cheaper than the mortgage, and we will say so.