Compare models on yearly income, not per-session rates. Multiply each model's session rate by the sessions you'd realistically fill under it: private pay averaged $159 per session to insurance's $111 in Heard's 2025 survey of nearly 2,000 therapists, but insurance caseloads fill faster, so a private-pay practice must stay roughly 70% as full to break even. Most therapists land on a hybrid.
Every few months some corner of therapist internet relitigates insurance vs. private pay, and both sides argue with averages. Averages don't run your practice. The decision comes down to one piece of arithmetic over three numbers you can actually get: your private-pay fee, your local contracted rates, and how full your calendar really stays without insurance referrals.
What's the real per-session gap between insurance and private pay?
About a third. In Heard's 2025 Financial State of Private Practice report, survey data from nearly 2,000 therapists across all 50 states, the average private-pay fee for individual therapy was $159, while the average insurance reimbursement was $111, or 36% less. Regional spread is wide on both sides: average session fees run from around $122 in the cheapest states to $227 in the most expensive, per SimplePractice's state-by-state data.
Your own gap may be smaller than the averages suggest. A well-negotiated 90837 rate with a strong payer can sit in the $130–150s, while an undifferentiated private practice in a saturated metro may struggle to fill at $159. Get your actual contracted rates from therapy reimbursement rates by insurance before plugging anything in.
Why does the vacancy rate decide everything?
Because an empty slot pays $0 under either model. Insurance panels come with a referral stream, since the payer's directory sends you clients, so paneled therapists typically run fuller calendars. Private pay pays half again more per hour, but you generate every referral yourself, and while you build that engine, slots sit empty at full price.
The break-even is simple: your insurance rate divided by your private fee tells you how full the private-pay practice must stay to match a full insurance practice. At $111 vs. $159, that's 70%. If you can't keep a private-pay calendar at least 70% as full as an insurance one, insurance wins on income, whatever the sticker rates say.
What does a worked example look like?
Take a therapist targeting 22 clinical hours a week, 46 working weeks a year, using the Heard averages:
| Model | Rate | Filled sessions/week | Annual gross |
|---|---|---|---|
| All insurance (90% full) | $111 | 20 | $102,120 |
| All private pay (60% full) | $159 | 13 | $95,059 |
| All private pay (75% full) | $159 | 16.5 | $120,681 |
| Hybrid (14 ins + 6 private) | mixed | 20 | $115,368 |
The hybrid row is 14 insurance sessions at $111 plus 6 private at $159. It beats all-insurance by $13,000 without requiring the referral engine that 75%-full private pay demands, which is why most practices land there: it is the highest number achievable with an ordinary referral flow.
Take-home runs lower than gross. Insurance income arrives with unpaid admin attached (eligibility checks, claim submission, denial rework, EOB reconciliation, the occasional records request), plus a clawback whenever a payer decides retroactively a session wasn't covered. Private pay's costs are marketing time and the awkwardness of holding a fee. Put your honest weekly hours for each into the worksheet at your own hourly value.
Leaping without the math is expensive. Leave two panels at once to go all private pay at $150 on the assumption that your insurance clients will convert, and you may find only a quarter of them do. Eight months near 40% capacity runs roughly $55,000 annualized, down from the $95,000 the paneled caseload earned, until rejoining the best-paying panel restores the floor. Same skills, same fee; only the vacancy rate moved.
If some clients have out-of-network benefits, a superbill lets them claim partial reimbursement while you stay private pay, which softens the fee objection without a contract. See the superbill template for therapists.
How do you run your own break-even?
YOUR THREE INPUTS
- Private-pay fee you can actually hold: $[FEE]
- Weighted average insurance rate (your real contracted 90837/90834 rates, weighted by which payers send you clients): $[RATE]
- Target clinical sessions per week: [TARGET]
STEP 1: Break-even fill rate $[RATE] / $[FEE] = [BE%] (Example: 111 / 159 = 70%. A private-pay practice must stay at least [BE%] as full as an insurance practice to match its income.)
STEP 2: Honest fill estimates Insurance model, sessions/week you'd fill: [I-FILL] (paneled therapists commonly run near capacity; use 85-95% of target) Private model, sessions/week you'd fill: [P-FILL] (be brutal: current inquiries per month x conversion, not hopes)
STEP 3: Annual gross (use 46 working weeks) Insurance: [I-FILL] x $[RATE] x 46 = $[A] Private: [P-FILL] x $[FEE] x 46 = $[B]
STEP 4: Subtract the invisible hours Insurance admin hours/week (eligibility, claims, denials, EOBs): [H] Insurance adjusted: $[A] - ([H] x your hourly value x 46) = $[A2] Private marketing hours/week: [M] Private adjusted: $[B] - ([M] x your hourly value x 46) = $[B2]
STEP 5: Test the hybrid [X] insurance sessions + [Y] private sessions (X+Y = realistic total): ([X] x $[RATE] + [Y] x $[FEE]) x 46 = $[C] Adjust X and Y until $[C] peaks. That ratio is your answer.
DECISION RULES
- $[B2] beats $[A2] only if Step 2's private fill was honest. Re-check it.
- If the hybrid peaks with 4-8 private slots, start there; raise Y as your waitlist grows.
- Recompute once a year, and whenever a payer changes your fee schedule.
How do you move between models without cratering income?
Gradually, and one panel at a time. Going insurance-to-private, cap new insurance intakes first and let attrition open slots you refill at your private fee, so income dips slowly instead of overnight and you learn your true private-pay demand at zero risk. Going private-to-insurance, get credentialed with your one or two best-paying local payers before you need them; the application months pass while your current caseload keeps paying.
The mistake in both directions is the cliff: resigning panels before the private referral engine is proven, or holding empty slots for months waiting on a panel you applied to late. Both are vacancy problems, and the worked example above shows vacancy moves annual income more than the per-session rate does.
What does the hybrid look like in practice?
Usually: stay on the one or two panels that pay you best and fill reliably, resign the ones paying bottom-of-the-band rates, and hold a fixed block of private-pay slots that grows as your reputation does. The panel choice does the heavy lifting; a $140 contract and a $95 contract are different businesses. If the math says drop a payer, do it cleanly: how to leave an insurance panel covers notice periods and client transitions.
And if what's stopping you from the insurance side is the paperwork rather than the rates, that part is delegable: services like paneled.ai exist to get the contracts into your name without the hours of forms and follow-up. The model decision stays yours; run the worksheet first.
Common questions
- Do therapists make more with insurance or private pay?
- Per session, private pay wins: therapists reported an average $159 private-pay fee versus $111 from insurance in Heard's 2025 practice survey. Per year, it depends on how full each model keeps you. An insurance caseload at 90% capacity routinely out-earns a private-pay caseload sitting at 60%, which is why the break-even math matters more than the sticker rates.
- What percentage of a therapy caseload is usually insurance vs private pay?
- Most sustainable solo practices end up hybrid, commonly anchoring on one or two well-paying insurance panels to keep the schedule full and reserving a block of private-pay slots at full fee. The right ratio falls out of your own numbers: your private fee, your local contracted rates, and how quickly you actually fill without insurance referrals.
- How much unpaid admin time does insurance billing add?
- Insurance adds work private pay doesn't have: eligibility checks before intake, claim submission, rejected-claim rework, EOB reconciliation, and occasional records requests. With an EHR that submits claims automatically the per-claim touch is minutes; without one it's substantially more. Estimate your own hours per week and price them into the comparison at your session rate.
- Should a new therapy practice start with insurance panels or private pay?
- Start from your referral reality. If you have a waitlist source (a niche, a strong network, an area short on therapists), private pay or a light hybrid can work from day one. If you'd be starting from zero visibility, insurance panels fill a calendar far faster, and you can rebalance toward private pay once you're full and can afford selectivity.
The paneled.ai team · Credentialing specialists. We file and track insurance credentialing applications for behavioral health providers every day — these guides come from what we see go wrong (and right) in real applications.