Google Ads are expensive for treatment centers structurally, and the click price is the symptom rather than the lever. The auction is where you notice the cost. It is almost never where the cost is created.
A program that drives its cost per click down for a quarter and watches its cost per admission rise has not been unlucky. It has optimised the number furthest from a patient in a bed.
Key takeaways
Five figures, and what each one should do to a plan.
| Finding | Figure | What it means for a program |
|---|---|---|
| Blended cost per admission | $618 | All channels together, across the portfolio |
| Paid media only cost per admission | $1,600 | 2.6 times the blended rate, run well |
| Ceiling on paid share of admissions | 39% | Forced by arithmetic, 618 divided by 1,600 |
| Two programs, same cost per lead | $1,667 and $10,000 | Hypothetical inputs, $200 a lead, converting at 12% and at 2% |
| Eight year lifetime blended average | $1,720 | Against $618 today, which is compounding rather than better bidding |
Those figures come from our own portfolio, 37,200+ tracked admissions and $64M+ in managed Google and Meta media since 2018, with definitions in the 2026 Behavioral Health Patient Acquisition Report. The channel distribution later is from five programs measured between 16 March and 11 September 2026.
In this article
- Why the click price is the wrong number to be angry about
- What actually sets the price of a behavioral health click
- Why published click cost benchmarks here are not usable
- Why cutting your cost per click usually raises your cost per admission
- Where the money actually leaks between the click and the admission
- Why two programs paying the same per lead pay different amounts per admission
- Why paid media stays expensive even when it is run perfectly
- Six things to run this week
- Questions operators ask
Why the click price is the wrong number to be angry about
It sits furthest from an admission. Between a click and a bed there are at least four more conversions, and the click is the only one the auction prices. Someone has to land on a matching page, call or submit, be answered by someone who can hold the conversation, verify benefits, then travel and admit. A twenty percent improvement in click cost improves the first step of five. A failure at any of the other four is total.
Click cost is the complaint we hear most and the lever we change least, because it is visible and updates daily. Cost per admission decides whether the channel is viable, and the two can move in opposite directions for a quarter while nobody notices.
What actually sets the price of a behavioral health click
Admission value, bidder concentration, search scarcity. Three structural facts, none of which an account manager can negotiate.
What a single admission is worth
An auction price is downstream of what the winner can afford to pay. In most categories a conversion is worth tens of dollars, so bids settle in cents. In behavioral health one admission carries a lifetime value in the thousands, frequently tens of thousands for a residential episode. Every bidder has done that arithmetic.
How few advertisers there really are
High intent treatment searches are contested by a small set of operators and networks with large, centrally managed budgets, several bidding nationally on the searches a single facility bids on locally. A few well funded bidders who all want the same impression is the most expensive auction structure there is.
How small the pool of real searches is
Most behavioral health search volume is research, definitions, symptom questions, insurance questions and job hunting. The subset where someone is ready to admit, or ready to place a family member, is narrow and does not expand because you raised your budget. Small fixed supply against concentrated demand is a seller's market.
What the certification requirement does and does not do
Google requires addiction treatment advertisers to hold third party certification, which limits how many qualify. It does not reduce competition among the ones who do, and those are the bidders with the budgets to compete hardest.
Why published click cost benchmarks here are not usable
Because they average incompatible things together. Read your own account instead.
A single average for addiction treatment has to blend residential with outpatient, self pay with insurance funded, brand with non brand, single facilities with national networks, and markets with wildly different competitive density. Those are not variations on one number, they are different auctions, and averaging them describes no real account.
There is also a sampling problem, because the accounts with the largest spend and the most competitive positions are least likely to feed a published benchmark. And the auction moves, since a competitor entering your market changes your costs inside a week. The only click cost that tells you anything is your own, by campaign and match type, read against admissions. How we read an account is on the Google Ads management.
Why cutting your cost per click usually raises your cost per admission
Because cheap clicks are cheap. Price in an auction is a proxy for intent, and the levers that cut what you pay per click mostly work by buying less intent.
- Lower your bids. You lose the top positions on the searches closest to admission and keep the rest. Click cost falls, admissions fall faster.
- Broaden your match types. Volume arrives, click cost drops, and the new traffic is research, students, job seekers and families gathering information. The denominator grows and the numerator does not.
- Add cheaper keywords. General and loosely related terms clear at lower prices because nobody competing for admissions wants them.
- Widen geography. Less contested markets are cheaper and often outside your realistic draw, so inquiries cost less and admit at a lower rate or not at all.
Each of those makes the dashboard look better and dilutes the population you are paying to reach. The most expensive clicks in a well run account are usually the most profitable, because that is where someone is ready.
Where the money actually leaks between the click and the admission
Downstream of the auction, almost always. When we audit an account producing admissions at several times the observed paid rate, the cause is rarely the bid.
- Calls that nobody answers. The largest recoverable loss we find. You pay full price for the click, the call rings into an unstaffed line after hours or at a shift change, and the inquiry is gone. Measure answered calls against placed calls first.
- Intake that responds too slowly. An inquiry's value decays in minutes. A family that submits at nine and hears back at four has already spoken to someone else. Intake response time sits downstream of every channel you buy.
- Broad match buying the wrong population. Research traffic, job seekers, students, clinicians and families gathering information for someone who will not admit for months. It all clicks, none of it admits.
- Conversion tracking that counts form fills. Optimise toward submissions and the algorithm finds the cheapest submissions, which are rarely the ones that admit. It optimises exactly what you tell it to.
- Keywords with no admission intent. Informational terms, insurance research terms, and terms for conditions you do not treat at levels of care you do not offer. These accumulate quietly over years.
- Leads bought at full price and never worked. Inquiries that get one attempt and no follow up. Families who were not ready on day one are the likeliest to admit in week three.
None of that is in the auction, and all of it is priced by the auction, because each failure happens after you have paid.
Why two programs paying the same per lead pay different amounts per admission
Because their conversion rates differ. The arithmetic is cost per lead divided by conversion rate. Take two programs paying the same price for the same volume, one converting inquiries to admissions at 12 percent and one at 2 percent. The inputs below are hypothetical, but the spread is the normal range here.
| Measure | Program A | Program B |
|---|---|---|
| Cost per lead | $200 | $200 |
| Lead to admission conversion | 12% | 2% |
| Leads needed per admission | About 8 | 50 |
| Media cost per admission | $1,667 | $10,000 |
Identical media prices and a sixfold difference in what an admission costs. Program A sits roughly at the observed paid rate of $1,600. Program B could cut its cost per lead by a quarter and still pay several times that, because the gap is not in the media. We treat cost per lead as a diagnostic rather than a goal.
Why paid media stays expensive even when it is run perfectly
2.6 times blended is the floor. Across the portfolio, blended cost per admission is $618 and paid media only is $1,600, so paid costs 2.6 times what everything together costs. That is not a badly run account, it is what a channel with a marginal cost costs when it is run well.
It also sets a hard ceiling. If blended is $618 and paid alone is $1,600, paid can account for at most 39 percent of admissions, because 618 divided by 1,600 is 0.386. A higher share would drag the blended figure above what it is. That is arithmetic, not opinion, and the remaining 61 percent has to come from somewhere that does not bill per click.
| Channel | Admissions | Share |
|---|---|---|
| Paid search and social | 307 | 47.7% |
| Referral partners | 109 | 17.0% |
| Organic search | 82 | 12.8% |
| Direct or untagged | 29 | 4.5% |
| AI labeled, source unverified | 116 | 18.0% |
| Total | 643 | 100% |
The five program sample runs paid at 47.7 percent of 643 admissions, above the ceiling, and those programs are still building the rest of the mix.
Our eight year lifetime average cost per admission is $1,720 against a current blended $618. That difference is not better bidding. It is compounding in channels with no marginal cost per admission, built while the paid account kept the beds full, which is the case for owning demand.
So the honest answer to why Google Ads are so expensive for treatment centers has two parts. Your account probably leaks between the click and the admission, which is worth fixing this month. Even fixed, the channel costs 2.6 times blended and caps near 39 percent of volume. The escape is not a cheaper click, it is a smaller share of your admissions depending on clicks at all.
Six things to run this week
None of these need a budget approval or new software.
1. Calculate your actual cost per admission
Last quarter's paid media spend divided by admissions your record credits to paid. Not leads, not form fills, not calls. If you cannot produce it today, that is the finding, and the most common one.
2. Count calls answered against calls placed
Pull a month of call data and compare it against what intake logged, by hour and day of week. The gap is admissions you have already paid for and did not take.
3. Time stamp your first response to every form fill
Submission time against first contact attempt, for thirty days. Read the median rather than the average, because one fast response hides ten slow ones. If the median is in hours, fix that before you touch a bid.
4. Read the search terms you actually paid for
Pull the search terms report, sort by cost, take the top one hundred real queries and mark each as admission intent, research intent or irrelevant. Most accounts find real money in the second two groups.
5. Check what your conversion actions optimise toward
Open the conversions list and read which action is primary. If it is a form fill, the platform has been buying cheap form fills on your instruction.
6. Write down your channel mix
Last quarter's admissions by source, from the admissions record, with one required intake field asking how the person heard about the program. If paid is well above 39 percent of census, the cost problem is a mix problem. Our attribution platform exists because that reconciliation is what most programs lack.
Questions operators ask
Why are Google Ads so expensive for treatment centers?
Because an admission carries a lifetime value in the thousands, a small number of well funded advertisers bid on the same narrow pool of high intent searches, and certification thins the field without softening competition among those who qualify. These clicks are expensive because they are valuable. But most programs paying too much per admission lose the money after the click, not at the auction. Across our portfolio paid costs $1,600 an admission against a $618 blended figure, 2.6 times.
How can a treatment center reduce its Google Ads costs?
Reduce cost per admission rather than cost per click. Measure your answered call rate and your median first response time, remove spend on terms with no admission intent, then check that conversion tracking counts something closer to an admission than a form fill. Lowering bids and broadening match types reduce the click price and raise the admission price.
What is a normal cost per click for addiction treatment advertising?
There is no usable published figure. Any average blends residential with outpatient, brand with non brand, and markets with very different competitive density, which describes no real account. Read your own account by campaign and match type instead.
Is a lower cost per click always better?
No, and frequently it is worse. Price in an auction tracks intent, so the methods that make clicks cheaper mostly work by buying a population less ready to admit.
What share of admissions should come from paid advertising?
At most 39 percent, which falls out of dividing $618 blended by $1,600 paid only. In a five program sample of 643 admissions, paid ran at 47.7 percent, which is a mix problem rather than an account problem.
Does certification make Google Ads cheaper for a rehab center?
No. Certification is what lets an addiction treatment advertiser enter the auction at all. It reduces how many advertisers qualify without reducing competition among those who do.
What we would look at first on your account
Your answered call rate, then your cost per admission. The first holds more recoverable admissions than anything in the bidding, and takes an afternoon. The second tells you whether the channel is genuinely expensive or has never been measured properly.
What we do day to day is on the services page, and results we have published are in our case studies.
Or run the six steps above and keep the answer. We would rather you knew your own numbers than took ours.
Related reading

Intent based marketing in behavioral health and what most campaigns miss
Google Ads for behavioral health practices and the 39 percent ceiling
