Original data · Engage Ads

2026 Georgia Medical Practice Advertising Benchmark

Methodology, stated up front

Most advertising "benchmarks" do not tell you where their numbers came from. Here is exactly where these came from, including what is weak about them.

Finding 1: In a fixed service area, cost per lead can double in six months without the ads getting worse

This is the finding worth the whole report, and it runs against the standard explanation.

An ENT and allergy practice ran a single Meta lead-generation campaign continuously for just over six months in a fixed suburban service area. Here is every month of it.

When cost per lead climbs like that, the usual diagnosis is creative fatigue: people have seen the ads too often, they have stopped responding, refresh the creative. In this campaign that diagnosis would have been wrong.

Between April and September:

• Click-through rate rose 35%, from 1.55% to 2.10%
• Leads per 1,000 people reached rose 18%, from 4.03 to 4.77
• Frequency stayed nearly flat, 2.16 to 2.34, so the same people were not being shown ads much more often
• Monthly budget was cut 30%, so the advertiser was not bidding up their own auction
• But the cost to reach 1,000 people rose 113%, from $51.52 to $109.55

The ads got better at their job on every measure of response. Cost per lead still rose 80%, and the arithmetic accounts for it completely:

The campaign did not run out of good advertising. It ran out of affordable people.

In a service area a patient will realistically drive to, the number of reachable adults is finite, and Meta charges progressively more as it works further into that finite pool.

What follows from this, practically: a local medical practice should expect cost per lead to rise over a long continuous campaign in one geography, and should not respond by firing the creative team. The levers that actually address it are geographic expansion, a different offer aimed at a different segment, or accepting a higher lead cost that the practice's economics can still carry. Refreshing creative alone does not buy back reach.

Finding 2: The offer moves cost per lead more than the targeting does

A chiropractic clinic in north metro Atlanta ran two different offers on Meta at the same time, to the same geography, in the same three and a half weeks.

Same practice, same city, same weeks, same budget scale. The low-barrier general offer produced leads at $15.85. The high-intent injury offer produced them at $56.60, 3.6× more. Counting both auto-accident variants together, including the one that spent $210 without producing a single lead, the combined auto-accident cost per lead was $109.16, nearly seven times the general offer.

This is worth stating plainly because it is routinely misread: the expensive offer is not the worse offer. An auto-accident case is worth far more to a chiropractic practice than a $49 exam. A cost per lead six times higher can still be the better investment. But it means cost per lead is close to useless as a comparison metric unless you know what the lead is worth, and practices that pick an agency or a campaign on cost per lead alone are optimizing the wrong number.

What we could not measure, and why it matters more than what we could

Both of these campaigns measured leads. Neither measured booked, attended, or paying patients, because the ad accounts carried no pixel, no dataset, and no offline conversion tracking. Meta recorded 676 completed lead forms for the ENT practice and knows nothing whatsoever about what happened to those 676 people afterwards.

We are publishing that gap rather than papering over it, for two reasons.

First, it is the honest thing to do, and any benchmark that quietly presents lead counts as though they were patients is misleading its readers.

Second, it is the central problem in medical practice advertising. Almost every number a practice sees from an agency, cost per lead, click-through rate, return on ad spend, all of it, is measured at the ad platform, which by construction cannot see the front desk, the schedule, or the deposit. A practice can be shown a genuinely excellent $16.65 cost per lead while the leads never convert, and nothing in the advertising report will reveal it.

The fix is not a better ad platform metric. It is connecting the practice's booking system back to the campaign, so the reported number is booked patients rather than form fills. Engage Ads reports in booked patients for this reason. The data in this report predates that tracking on these two accounts, and we are not going to pretend otherwise.

We expect to publish cost per booked patient in the next edition of this benchmark. Until then, treat every cost-per-lead figure here, ours and everyone else's, as the top of the funnel and nothing more.

How this compares to published industry figures

To make our two data points useful, they need something to sit against. Below are the published benchmarks we could actually verify, each with its publisher, sample and period. We went looking for figures covering the same verticals, and mostly did not find them, which is itself the most useful thing in this section.

Meta, lead-generation objective: the like-for-like comparison

Our data is Meta lead forms, so this is the only fair comparison table.

Sources: WordStream Facebook Ads Benchmarks 2025, published Sep 15 2025, medians. LocaliQ Healthcare Advertising Benchmarks. Superads healthcare cost-per-lead.

Both of our figures come in far below every published Meta healthcare comparison. The ENT practice's $16.65 is 65% below the closest like-for-like figure, physicians and surgeons at $47.47, and below even the all-industries median of $27.66.

Google, for context (different platform, not comparable to our numbers)

Sources: LocaliQ Healthcare Search Advertising Benchmarks, published Jan 13 2026, 3,542 US campaigns, Oct 2024–Sep 2025. WordStream 2026 Google Ads Benchmarks, published May 19 2026, 13,474 US campaigns, Apr 2025–Mar 2026. AmSpa, "The Marketing Investment Gap," Jan 9 2026, reporting Growth99's survey of 81 practices.

One cross-check that does line up: our chiropractic auto-accident figure of $56.60 sits almost exactly on LocaliQ's published chiropractic search cost per lead of $57.97, though that is a different platform and the data was collected in 2020–21, so treat the agreement as a coincidence worth noting rather than a validation.

Where published benchmarks simply do not exist

We looked hard for these and could not find them from any platform, ad-tech aggregator, trade association or academic source:

• ENT and allergy advertising costs. No published benchmark exists. LocaliQ's 16-specialty healthcare dataset explicitly excludes ENT and allergy. Everything else we found was a single-client agency case study with no methodology.
• Current chiropractic advertising costs. Nothing has been published since LocaliQ's 2020–21 collection window, which pre-dates the iOS 14.5 tracking changes.
• Med spa costs measured from ad accounts. The only figures available come from an 81-respondent survey, not from platform data.
• Meta cost per thousand impressions broken out by healthcare specialty. WordStream publishes no CPM figures at all in its Facebook benchmarks.
• Any Georgia or Southeast-specific cut. Every source we found is US national.

That is the honest case for this report existing. Our sample is two practices; theirs are thousands of campaigns. But for ENT, for current chiropractic, and for local Georgia performance, two documented practices is more than anyone else has published.

Where the published sources disagree with each other

Worth knowing before you trust any single benchmark, including ours:

• Dental cost per lead on Google differs by 16% between two reports from the same publisher family. WordStream's 2026 figure is $72.97; LocaliQ's healthcare report says $84.77 for general dentistry. Different samples, different windows, same parent company.
• Healthcare cost per lead on Meta differs by more than 2× between WordStream's $47.47 for physicians and surgeons and Superads' $103.52 for healthcare in July 2026.
• Med spa cost per lead is reported at $39 by AmSpa citing Growth99's survey, against WordStream's measured $51.42 for beauty and personal care on Meta.

Two figures we deliberately left out. A "$5 to $10 med spa cost per lead" range and a claim that healthcare Meta CPM rose 70% between January 2025 and January 2026 both circulate widely, and neither could be traced to anyone who actually measured them. They are not in this report.

Speed to lead: what the research actually says, and who actually said it

Contacting a lead quickly matters enormously, and the evidence for that is strong. It is also misattributed almost everywhere, including by marketers who should know better.

We have not measured our own speed-to-lead figure, so we are not publishing one. What follows is other people's research, correctly attributed.

The origin of the "five-minute rule" is MIT, not Harvard

The famous multipliers come from research by Professor James Oldroyd using InsideSales.com data: three years of data from six companies, more than 15,000 leads and over 100,000 call attempts, published through the Lead Response Management study.

• Odds of contacting a lead drop 100× when the first call comes at 30 minutes rather than 5 minutes
• Odds of qualifying a lead drop 21× over that same gap
• Qualification odds fall roughly 4× between minute 5 and minute 10
• After about 20 hours, additional call attempts become counterproductive

Lead Response Management study, Oldroyd/MIT with InsideSales.com, leadresponsemanagement.org

Harvard Business Review published a different study, with different numbers

In March 2011, Oldroyd, Kristina McElheran and David Elkington audited 2,241 US companies by submitting a test lead through each company's website.

• 37% responded within an hour, 16% within 1 to 24 hours, 24% took longer than a day, and 23% never responded at all
• Among companies that did respond within 30 days, average response time was 42 hours
• Firms responding within an hour were roughly 7× more likely to qualify the lead than those responding an hour later, and more than 60× more likely than those waiting a day

Oldroyd, McElheran and Elkington, "The Short Life of Online Sales Leads," Harvard Business Review, March 2011

Both studies are cross-industry business-to-business sales data, not healthcare. We are not aware of any published, methodologically documented speed-to-lead study specific to medical practices. Several healthcare marketing sites publish precise-sounding figures, a commonly repeated pair being that responding within 21 minutes lifts lead-to-consultation conversion by 65% and that the average practice takes 47 hours to respond, but we could not trace either to a stated sample, method or measuring organization, so we are not repeating them as fact.

The honest position: the direction of the effect is very well established across industries and the magnitude is large. Whether it is the same magnitude at a medical front desk is, as far as we can find, unmeasured. We intend to measure it on our own accounts and publish it here.

Does anyone else's data show audience saturation?

Our finding that reach cost rose 113% in a fixed service area has one useful independent analogue. Sotrender analyzed more than 20,000 Meta ad sets and found that median CPM rose 115% between ad sets with under 10% audience saturation and those above 90%, and that "after reaching 90% of the target audience, the average CPM doubles."

Sotrender, "Audience saturation," analysis by Maciej Baranowski, published September 26 2025; the underlying data was collected in Q1 2022.

Their +115% and our +113% are strikingly close, and they were measured completely differently: theirs across 20,000 ad sets by saturation level, ours across one campaign over six months in a fixed geography.

Frequently asked questions