Watch five numbers every week: cost per admission, conversion at each funnel stage, length of stay, payer mix with net revenue per admission, and referral source concentration. For scale: the largest public agency dataset we found put paid-media cost per admission at $16,608 in 2025, with 2.4% of leads admitting (Webserv 2025).
What we see running campaigns for detox and residential is lower: about $2,000 to $5,000 per admission in-network and $7,000 to $15,000 out-of-network.
Written for owners and CEOs. Every benchmark below links to its source or is labeled as our experience or our math. Where no good public benchmark exists, we say so and tell you to set your own baseline.
Why aren't total inquiries and admits enough?
"450 inquiries and 62 admits last month" doesn't tell you what it cost, what each admit was worth, or how exposed you are to one channel. The five numbers below split those totals into parts you can act on. Track them weekly. When one moves, you know where to look.
The rule: if a number doesn't change what you do on Monday morning, it's a report, not a KPI.
KPI #1: Cost per admission
Formula: total marketing cost ÷ admissions in the same period.
Include everything it takes to get an admit through the door: ad spend, agency and SEO fees, directory fees, call tracking, CRM and admissions staff time if you want a fully loaded number. Count admissions who arrived and started treatment, not scheduled no-shows. Break it out by channel and by level of care. A blended number hides which channel is losing money.
Benchmarks
| Source | Cost per admission | Notes |
|---|---|---|
| Our experience: Google and directories, in-network | $2,000 to $5,000 | Detox and residential |
| Our experience: Google and directories, out-of-network | $7,000 to $15,000 | Detox and residential |
| Our experience: SEO | A few hundred dollars to $2,000 to $5,000 | Detox and residential |
| Webserv 2025 | $16,608 | Paid media, all its clients, 994 admits |
| Webserv 2024 | $14,987 | Clients with 6+ months of PPC, outliers excluded |
PHP and IOP usually cost less per admission. In our experience the range varies too much by region and payer to publish one number.
Our math, from SEC filings: American Addiction Centers disclosed its ad and marketing expense and its admissions. $15.683M ÷ 4,728 admissions = about $3,317 per admission in 2014. $12.315M ÷ 12,299 residential admissions = about $1,001 in 2017 (AAC 2017 10-K, 2015 10-K). Treat these as a floor. The ad line leaves out sales and call center staff, and AAC owned directories like Rehabs.com.
When cost per admission rises, fix the funnel before you add spend:
- Landing page. Is the phone number and VOB form visible on a phone screen without scrolling?
- Response time. HBR's audit of 2,241 US firms found those that contacted a web lead within an hour were nearly 7 times as likely to qualify it as those that waited longer (HBR, 2011).
- VOB handoff. How long from first contact to a completed VOB?
- Keywords. Are you paying for research searches that never call?
KPI #2: Conversion at each funnel stage
Formula: for one cohort of leads, the share that reaches each stage: lead, VOB, viable VOB, admit.
One "lead-to-admit" number can't tell you whether marketing or admissions is the problem. Stage rates can. Webserv's published funnel for paid media:
| Stage | 2025 | 2024 |
|---|---|---|
| Lead to VOB | 15.9% | 18.9% |
| VOB to viable VOB | 42.8% | 43.8% |
| Viable VOB to admit | 35.9% | 37.7% |
| Lead to admit | 2.4% | 2.7% (our math: 445 ÷ 16,555) |
Sources: 2025, 2024. The 2024 set excluded outliers and the 2025 set did not, so the years aren't a clean comparison.
How to read your own funnel:
- Low lead to VOB: Webserv says this gap is "almost always a lead quality or speed-to-contact issue."
- Low VOB to viable: the benefits don't fit your program. Look at keywords, geography and payer targeting.
- Low viable to admit: an admissions problem. Listen to the calls.
Some agency pages quote 15% to 25% from inquiry to admission with no published data behind it (for one). Check what they count as a lead before you compare.
KPI #3: Length of stay
Formula: total treatment days for patients discharged in the period ÷ number of discharges. Report it by level of care.
AAC listed the drivers of revenue per admission as "in-network or out-of-network insurance coverage, the level of care for which the patient is receiving, and the average length of stay" (AAC 2017 10-K).
Benchmarks: SAMHSA TEDS 2023 median length of stay
| Service type | Median days |
|---|---|
| Residential, short-term | 21 |
| Residential, long-term | 36 |
| Hospital residential | 7 |
| Detox, free-standing residential | 5 |
| Detox, hospital inpatient | 4 |
| Intensive outpatient | 40 |
| Non-intensive outpatient | 53 |
Source: TEDS 2023 annual report, Figure 4.B.6. TEDS mostly covers publicly funded treatment reported by state agencies, so a private-pay or commercial program may run longer. Use it as a reference point, then track your own trend.
Our math: revenue lost = days lost × net daily revenue × admissions. AAC reported average daily residential revenue of $873 in 2017. At that rate, losing 2 days per stay costs 2 × $873 = $1,746 per admission. For a center with a hypothetical 200 admissions a year, that's $349,200. Plug in your own daily rate.
Watch a 4-week rolling average by level of care. When it drops, check insurance authorizations, discharges against medical advice and step-down timing.
KPI #4: Payer mix and net revenue per admission
Formula: share of admissions by payer × average net collections per admission for that payer. Add them up for total revenue.
Two centers with the same admission count can have very different revenue. AAC said its average daily residential revenue from in-network beds "is generally lower" than from out-of-network beds. It also reported that in-network beds made up 16.4% of client revenue in 2016, up from 6.3% in 2015 (AAC 2017 10-K).
We don't publish net revenue per admission by payer. It depends on your contracts, your state and your billing team. Pull it from your own collections data. For context, SAMHSA's 2024 survey found that among substance use facilities taking payment, 89.9% accept cash, 78.3% accept private insurance, 77.8% accept Medicaid and 52.6% accept Medicare (N-SUMHSS 2024).
Track payer mix weekly and alert on drift. If marketing moves to a cheaper channel, check that the admits it brings pay the same. See PPO vs cash-pay marketing.
Want us to audit your reporting setup?
We audit rehab reporting stacks and find the numbers that are missing, defined wrong, or reviewed too slowly to catch a leak. Free 45-minute executive audit.
Book a free auditKPI #5: Referral source concentration
Formula: share of admissions from each source.
This one measures risk, not revenue. Track each source on its own:
- Google Ads
- Meta Ads
- SEO and organic search
- Google Business Profile
- Directories
- Alumni referrals
- Professional referrals: therapists, doctors, interventionists
- Hospitals and insurance case managers
- Direct and word of mouth
Channels do switch off. AAC's CEO reported "an initial 30% drop in call volume resulting in sharply decreased admissions" in the second half of 2018 (AAC 8-K, March 2019). Google warns at least 7 days before it suspends an account for an addiction services policy violation (Google Ads policy). That's not much time to replace your largest source.
There's no published standard for a safe share. Set your own ceiling for any one source and build a plan when you cross it. Start slower channels early. Google has said an SEO will usually need "four months to a year" to make changes and see the benefit (Google's Maile Ohye, via The SEM Post).
Other numbers worth tracking
- Days from first inquiry to admission. Funnel speed, by channel.
- Approved-to-arrived rate. One CRM vendor says most programs lose 20% to 40% of approved admissions before arrival (Navix Health, vendor claim). Measure yours.
- Margin per admission. Our math: net revenue per admission − direct clinical cost − cost per admission.
- Outcomes. Whatever your accreditor and payers ask for. Track it the same way every month.
The review cadence
Weekly (Monday, 30 minutes)
- All five numbers, week over week
- Payer mix and referral source shares
- Anything off trend assigned to an owner
Monthly
- Cohort view: leads to admits by week of first contact
- Cost per admission by channel and level of care
- Length of stay distribution, not just the average
Quarterly
- All numbers against annual targets
- Decisions: hire, invest, pull back, add levels of care
Where does the data live?
- EHR: admissions, length of stay, discharges.
- CRM and call tracking: leads, sources, VOBs, the admissions pipeline.
- Billing: payer, collections, net revenue.
If these don't connect, you can't see cost per admission by source or net revenue by payer without spreadsheets. When you pipe patient data into a dashboard, remember the rules. HIPAA lets you share protected health information with a business associate only with "satisfactory assurance" it will be safeguarded (45 CFR 164.502). Substance use records under 42 CFR Part 2 "may be used or disclosed only as permitted" by that rule (42 CFR 2.13). Ask your compliance lead before you connect systems.
Common mistakes
- Watching totals only. Total leads and cost per click don't tell you what an admit cost or what it was worth.
- No cohorts. A lead from this week may admit next month. Track by week of first contact.
- Blended cost per admission. Split it by channel or you can't see which one to cut.
- Quarterly-only reviews. Too slow to catch a channel switching off.
- Three versions of the truth. Marketing, admissions and finance each have a number. Fix the data pipe first.
- Paying for admits. Federal law (EKRA, 18 U.S.C. § 220) bars paying to induce referrals. Track your own cost per admission. Never pay vendors or staff per admission.
What good looks like
- All five numbers updated by Monday morning without manual work
- Cost per admission by channel and level of care
- Funnel stage rates by source and by admissions rep
- Length of stay as a 4-week rolling average by level of care
- Payer mix tracked weekly with a drift alert
- A written ceiling for any one referral source
Frequently asked questions
What is a good cost per admission for a rehab?
It depends on payer and level of care. What we see running campaigns for detox and residential: about $2,000 to $5,000 per admission in-network from Google and directories, $7,000 to $15,000 out-of-network, and a few hundred dollars to $2,000 to $5,000 from SEO. Webserv's 2025 dataset put paid-media cost per admission at $16,608 across 994 admits. Fix the funnel before you add spend.
What is a healthy lead-to-admit rate?
Track each stage, not one rate. Webserv's 2025 paid-media data: 15.9% of leads reached a VOB, 42.8% of VOBs were viable, 35.9% of viable VOBs admitted, and 2.4% of all leads admitted. A low lead-to-VOB rate points to lead quality or speed to contact. A low viable-to-admit rate points to admissions.
What length of stay should I benchmark against?
SAMHSA's TEDS 2023 report lists median stays of 21 days for short-term residential, 36 days for long-term residential, 5 days for free-standing residential detox and 40 days for intensive outpatient. TEDS mostly covers publicly funded treatment, so use it as a reference point and track your own 4-week trend by level of care.
How much revenue do I lose if length of stay drops by 2 days?
Our math: days lost × net daily revenue × admissions. AAC reported average daily residential revenue of $873 in 2017. At that rate, 2 fewer days cost $1,746 per admission, or $349,200 for a hypothetical 200 admissions a year. Use your own daily rate.
Why does payer mix matter so much?
Revenue per admission depends on in-network or out-of-network coverage, level of care and length of stay, as AAC listed in its 2017 10-K. Two centers with the same admission count can earn very different revenue. Pull net revenue per admission by payer from your own collections data and track the mix weekly.
What's the risk if one referral source drives most admits?
If it switches off, census drops. AAC's CEO reported an initial 30% drop in call volume that sharply cut admissions in late 2018. Google gives at least 7 days of warning before suspending an account for an addiction services policy violation. Set your own ceiling for any one source and start slower channels like SEO early.
How often should I review these numbers?
Weekly for all five, a monthly cohort review, and a quarterly review against annual targets.