Should You Start a Therapy Group

A therapy group is not simply an individual session with more people in the room. It has its own clinical purpose, confidentiality limits, attendance pattern, documentation workload and financial model, and treating it as a cheaper version of individual work is where most of the planning goes wrong.

“Are therapy groups profitable” is too broad to answer. The useful question is whether one specific group, at your rate and the attendance you can realistically hold, returns enough per hour of your time to justify running it.

A group is worth planning when four things are true. You can name the population and the clinical purpose in a sentence. You can screen and support the members it would attract. You can carry the documentation and administration it creates. And your modeled attendance clears your own hourly target. The rest of this article is how to check the fourth one, and why the first three decide whether the fourth matters.

This is planning guidance from a marketing agency. It is not clinical, legal, billing, tax or compliance advice, and it does not determine what you can bill or how you should document. Confirm those with your payer, your licensing board, and qualified local guidance.

Group Therapy and Group Practice Are Different Decisions

Search for whether groups are worth it and most of what comes back is about group practice. Hiring clinicians, employee versus contractor, revenue splits, practice overhead. That is a different business decision with different math.

This article is about one clinician running one recurring group. If your question is about adding clinicians, the therapy cash flow calculator models that directly, including break-even sessions per clinician and the point where a group practice covers its overhead.

A Full Group Can Gross More Than One Individual Session

The usual assumption is reasonable. If each member pays less than an individual client, the group should earn less. The arithmetic runs differently, and the reason has nothing to do with your fee.

The first half is true.

Published government data

18.2%

one group claim against one 60-minute individual claim, Medicare comparison

The second half is false, and it is false for a mechanical reason. For covered group psychotherapy, payment is generally made per participating patient rather than once for the whole room. Each attending member is a separate claim, and payer rules and rates differ.

Published government data

1.09x

six attending members against one individual session, same Medicare comparison

At six attending members, this Medicare example grosses slightly more than the individual session it replaced. That is a fact about gross revenue and not a verdict on the group. Anyone who has run a group and found it exhausting rather than lucrative is describing something real, and the gross revenue is not where it lives.

Two cautions on those figures before you use them. They come from one administratively set national schedule on one date, they are adjusted geographically in practice, and commercial and cash-pay rates are set independently of all of it. They tell you which direction the arithmetic runs. They do not tell you what you will be paid.

The Work Outside the Session Changes the Math

The clinical hour is one hour. The work attached to it is not.

Group psychotherapy claims are submitted per participating patient, and each patient needs documentation specific to them. Their own diagnosis, their own treatment-plan connection, their own participation and response. A shared group-level record describing the session is useful and does not replace it. Six attending members means six notes.

That mechanism runs in the opposite direction from relief. Revenue scales with attendance. So does documentation. The fuller the room, the more notes it produces.

Screening runs ahead of all of it. A pre-group conversation with each prospective member is standard practice, and not every person who screens will enroll, so the screening work usually exceeds the number of seats you fill.

Here is what that adds up to for one weekly group. The numbers below are one therapist’s planning assumptions, including a nine-minute estimate for each participant note. No independent research establishes a standard per-participant documentation time, and the figures published for it come from companies selling documentation software. Replace it with what your own workflow takes.

Two things about this ledger matter later. The recruitment and screening hours are fixed per cohort, so they get cheaper per session the longer the cohort runs. The note hours are the opposite, and they never get cheaper.

What Your Group Is Worth at Different Attendance Levels

Higher attendance improves the result, but not every attendance level clears the bar.

Input or result	3 attending	4 attending	6 attending	8 attending
What we assumed
Members attending	3	4	6	8
Collected per attending member	$60	$60	$60	$60
What the model produces
Gross for the session	$180.00	$240.00	$360.00	$480.00
Revenue remaining after direct costs	$132.10	$190.30	$306.70	$423.10
Total therapist hours, session and everything around it	3.10	3.25	3.55	3.85
Effective return per therapist hour	$42.61	$58.55	$86.39	$109.90
Meets this therapist's $110 target?	No	No	No	Almost

This is an illustrative planning model, not a benchmark. At $60 collected per attendee, even a full room of eight only nearly reaches this therapist's $110 target once screening, documentation, marketing and direct costs are counted. Three attending is below the minimum she set to run the session at all.

Illustrative worked example · one weekly 90-minute group, 12-session cohort, $60 collected per attending member, $30 room, 3% processing, $150 marketing · figures derived, not observed

The row that matters is the last one, and the reason it climbs so steeply is the reason groups feel like they should work.

Every additional member brings in another fee and costs one more note. Most of that fee lands in what the group contributes, and the note is the only part of the workload that grows with them.

The average hour looks poor because 2.65 hours are already spent before the first person arrives. A fuller group improves that average, and the group still has to clear your bar across the whole workload.

Once the group is clinically sound, consistent attendance usually moves the arithmetic more than a small price increase, because an added member changes the denominator barely and the numerator a lot. Neither one should override fit, safety, or what the group is for.

Which is exactly the pressure to name out loud, because this article created it. A page that shows you what one more attendee is worth has just made a case for filling seats, and an underfilled group is where that case turns into accepting someone who does not fit, keeping a group running that is not working, or not referring out a person who needs more than the format can give.

Decide those in advance, while the room is still hypothetical and the arithmetic is not sitting in front of you. Minimum enrollment, the largest size the group can hold well, what fit means for this group, when you pause, and when you refer. Those are clinical decisions that happen to have financial consequences, and the order matters: the seat math tells you whether to run the group, and it does not get a vote on who is in it.

What One Absence Costs

A no-show in individual work costs you the slot, and the hour is free.

A no-show in a group costs you the fee while the session runs anyway. The room is booked, the preparation is done, the shared record still gets written, and your hour is spent. In the example above, each absence removes $58.20 of net revenue and gives back only the minutes that note would have taken.

Individual work has variable revenue and variable cost. A group has variable revenue and fixed cost. That asymmetry is the financial shape of the format, and it is why attendance policy in a group is a different problem from a cancellation policy in individual work.

It also means a group can look healthy on its enrollment list and still run below its intended size in a given week.

Peer-reviewed research

74.9%

group sessions that ran below their ideal size

Those are UK trial cohorts, recruited and supported under protocol, so treat the direction as informative and the number as belonging to somebody else’s setting. Plan around the attendance you can hold, not the roster you enrolled.

Use Your Own Collections to Find Your Rate

One published analysis does give a defensible cross-state picture, and it is the closest thing to an answer that exists.

Peer-reviewed research

$23.33 against $110.86

median state Medicaid payment, group against 60-minute individual

Read the scope before you use it. Those are Medicaid fee-for-service rates paid to psychiatrists in 2022, and the authors say plainly that they could not draw conclusions about services delivered by mental health professionals who are not psychiatrists. The interquartile ranges were $20.68 to $28.17 for the group code and $95.19 to $134.57 for the individual one, so states sit a long way apart.

In that psychiatrist fee-schedule analysis the median group payment was about a fifth of the median 60-minute individual payment, with wide variation across states. Our own reading of two state schedules put Texas Medicaid at 18.2% and Alaska at 40.0%, which is the same story with a wide spread.

Vendor-published figures for the group code are less reliable than that. Reviewing the pages that rank for it, the published numbers for the same code in the same year run from about $30 to about $65, and one of the widest ranges sits on a page whose own disclaimer calls its ranges illustrative. Published ranges like these are broad, incomplete, and not tied to your credential, your contract or your setting.

You may also meet a figure saying a group needs about seven members to break even. It comes from a hospital psychology department, the word in the original is “credit,” and it compares against two 45-minute sessions rather than one. It is describing a different workload in a different setting, so it is not a substitute for your own arithmetic.

None of this is a lookup. Your rate depends on your payer, your contract, your state, your license and your setting. Pull it from your own remittance data, the same way you would find your average collected revenue per individual session.

Decide What Kind of Group This Is

Decide this before pricing, promotion or enrollment. The service, rather than the price, determines the clinical and administrative structure it requires.

Format	What it is for	What it carries
Free psychoeducational workshop	Teaching a topic or skill	Usually not psychotherapy. Do not present it as treatment
Free peer-support or community group	Connection and mutual support	Clear boundaries, facilitator role, safety plan, privacy expectations
Paid private-pay therapy group	Psychotherapy with a clinical purpose	Screening, group consent, clinical records, payment policy
Insurance-billed therapy group	Covered psychotherapy	Payer-specific coverage, authorization, coding and individualized documentation

Charging a fee does not convert education into therapy, and offering something free does not exempt therapy from the structure therapy requires. The service decides the label.

A free introductory group deserves a specific warning, because it is advice you will find elsewhere. Running a free first session as a trial, the model gyms use, collapses the line between an outreach event and a clinical service. If it is therapy, it needs consent, screening, documentation and a safety plan whether or not anyone paid. If it is education or support, describe it that way and do not promise treatment.

A free therapy group can carry the same screening, consent, documentation, privacy and crisis-planning obligations as a paid one. Before offering one as a way to generate referrals, be clear about what clinical responsibility you are taking on.

Because we help practices market their services, this distinction is one to state plainly: marketing should not decide whether a clinical group exists, or who belongs in it.

A free group can still be a reasonable choice. Model it honestly. Direct session revenue is zero, while the screening, preparation, documentation and crisis-planning work may all still be there. Any return comes from the purpose you set for it, whether that is education, access, community connection or referrals.

Screen for Fit Instead of a Blanket Exclusion List

Screening is an individual conversation before enrollment, not a sign-up form. How long it takes depends on the group, the person’s needs, and whether you already know them. Four things get assessed:

• Fit. Does this person’s concern and goal match what the group is for?
• Readiness. Can they participate with peers, receive and respond to feedback, and work within the group agreements?
• Safety and stability. Can they participate without the group becoming unsafe or unable to meet their needs?
• Logistics. Can they attend consistently, pay, and use the platform privately if it is remote?

Screening should lead to more than a yes or no: ready to enroll, potentially appropriate with preparation, better suited to another group or to individual therapy, or in need of a different level of care now.

You will find articles listing diagnoses and presentations that exclude someone from group work. Those lists are a mistake. Groups exist that are designed specifically for suicidality, psychosis and substance use, and the literature on group selection warns that excluding for low motivation, hostility or limited social skills denies treatment to people who may most need it. Suitability is relative to a particular group’s purpose, structure and support, which is a judgment your screening conversation exists to make.

The AGPA practice guidelines and SAMHSA’s TIP 41 both cover placement and preparation properly, and both are free to read.

One more thing belongs in that conversation. Confidentiality binds you and your practice. It does not bind the other members in the same way. A written group agreement and a no-recording rule are standard and neither guarantees what someone says outside the room, so say so during screening rather than after.

Set a Minimum Attendance Number Before You Market

You will find conflicting advice about the ideal group size, and no universal answer behind it. The right number depends on your payer’s rules, the treatment model, how much support the members need, the format, and how many people you can hold safely in one room.

Instead of one ideal number, check three constraints:

• What your payer caps at, if you are billing insurance.
• What the room or platform seats, and what the format can hold clinically.
• What your own arithmetic needs, which is the minimum average attendance that clears your hourly bar.

The third one is the only number in this article you cannot get from anywhere else, and it moves with your fee, your session length, and how much non-session work the group requires.

Choose a Narrow First Group

For a first group, narrower is easier in every direction. Pick a concern you already treat often, a population your existing referrals can reach, and a format you can explain in one sentence.

“Anxiety group” is broad. “A six-week skills group for adults managing social anxiety” gives a prospective member and a referring colleague a reason to choose it over the other options in front of them.

A narrow promise also makes screening easier, because you already know what the group is for and who is likely to benefit. That is the same clarity that tells you when individual therapy, a different group, or another level of support is the better answer for someone.

Build a Group You Can Repeat Without Rebuilding It

The hours ledger points at its own fix. Recruitment, screening, consent materials and session planning take the most work before the first cohort, and a group designed to run again reuses most of it instead of rebuilding it. A group built to run once is expensive every time.

That changes what a first group should look like.

For a first run, a closed cohort with a defined start and end date is easier to explain, screen, document and evaluate than an open-ended group. Set your minimum enrollment and your clinical capacity before promotion begins, so that when enrollment closes you are making a decision you already defined rather than one the numbers are making for you.

Screen individually instead of enrolling first-come. That is where fit gets decided, and it is also where you find out whether the promise you wrote is the one people are responding to.

Then build the paperwork once. One consent packet, one confidentiality agreement, one session outline, one participant-note template, one attendance process. Every hour spent on those before the first cohort is an hour you do not spend during the twelfth session of the second one.

Treat the first run as a pilot internally and as a real group externally. Internally it is a test of whether demand exists, where referrals come from, what attendance looks like, and how long the documentation takes. Externally it needs everything a group needs: consent, screening, records, a crisis plan, and a clear answer for what happens if enrollment falls short.

Review it when the cohort ends. Attendance against enrollment, where each member came from, the hours you spent, and what the group contributed against your bar. That review is what turns one group into a repeatable one, and it is the input the second cohort’s arithmetic depends on.

One thing not to treat as a pricing problem. Attendance responds to whether members understand why they are there, feel safe enough to take part, and can see the structure working. Lowering the fee does not fix a group whose purpose is vague, and it removes the commitment that keeps a seat filled.

If you have not run a group before, the training is a separate investment from the setup. The American Group Psychotherapy Association runs foundational courses, experiential workshops and local affiliates, and consultation with someone who runs the kind of group you are planning is worth more than any template.

Before You Launch

A checklist, not a score. Nothing here is totaled and none of it is a determination.

• Have you decided whether this is psychotherapy, psychoeducation or support, and does the public description match?
• Does each enrolled member have an individual clinical reason to be in this group?
• Have you confirmed coverage, authorization and documentation requirements with each payer you plan to bill?
• Do you have group-specific informed consent and a confidentiality agreement?
• Can your records system hold a shared session record plus individualized notes per member?
• Have you put screening, documentation, billing follow-up, attendance chasing and seat refill into the calendar as hours?
• Do you have a protocol for fit, safety, crisis response and referral out?

When a Group Earns Its Place Anyway

The example in this article loses. Many will. That is a real finding, and the conditions where groups win are specific enough to check against your own situation.

A repeat cohort with a waitlist changes the arithmetic most, because recruitment and screening are the fixed costs and a second cohort inherits the work of the first. A shorter session raises the return per hour directly. Admin that does not scale with headcount helps. So does a fee that is a large share of your individual fee rather than a small one.

Peer-reviewed research

g = -0.01

group against individual therapy, same treatment and dose

When comparable treatments, patients and doses are compared, the two formats have shown statistically indistinguishable outcomes overall. That does not make them interchangeable for every person or every clinical goal, and it cannot explain why groups are under-supplied. What it does mean is that the operational and financial barriers deserve attention alongside clinical fit.

And the return may not be financial. A group can be a lower-cost entry point into your practice, a way for people to begin before committing to individual work, or a service line that serves people your calendar could not otherwise reach. A group does not have to replace an individual caseload to earn its place in a practice.

What it does have to do is survive an honest count of the hours. If you run yours through the arithmetic above and it clears your bar, you have a real answer instead of an assumption. If it does not clear, you have learned that before committing yourself and prospective members to a cohort.