Software is a partnership. It should not cost a C-suite salary every year.

Look at the annual software line item at a lot of treatment centers and it reads like an executive's salary. Worse, it climbs every renewal, often because features you assumed were included get unbundled and re-priced a year or two after you signed.

That number deserves the same scrutiny you'd give a six-figure hire. What did you actually get for it, and is it still working for you now.

The real cost is bigger than the sticker

Legacy clinical software has never been cheap to own. A frequently cited study put implementation at roughly $162,000, plus about $85,500 in first-year maintenance, for a small practice (Health Affairs, 2011). Those figures are more than a decade old, but the pattern they describe has not gone away: behavioral health enterprise pricing is still largely quote only, which makes side by side comparison hard by design.

The incumbents also tend to stack modules. The EMR is the entry price, and the CRM, the RCM, telehealth, and now AI features each carry a separate fee. Add implementation hours, interfaces, data migration, custom reports, training, and support tiers, and the number you signed for at go live is rarely the number you're paying by year three.

That gap matters more now than it used to, because the workforce backing your operation is thinner than it was. As of December 2, 2025, roughly 137 million Americans, about 40% of the U.S. population, lived in a Mental Health Professional Shortage Area, per the Health Resources and Services Administration (HRSA, 2025). HRSA projects the gap will widen: by 2038 it forecasts shortages of approximately 77,050 addiction counselors, 99,780 mental-health counselors, 99,840 psychologists, and 43,810 psychiatrists (HRSA workforce projections). Meanwhile, an estimated 62 million U.S. adults, 23% of adults, had a mental illness in 2024, and 48% of those adults went untreated (HRSA, 2025). Demand is high, staff are scarce, and every dollar and every hour your software wastes is a dollar and an hour you don't have to spare.

Opaque pricing is leverage, not an accident

A vendor who won't publish pricing and who re-prices at renewal is exercising leverage. The moment your data and your workflows are inside the system, the cost of saying no goes up, and the vendor knows it. The steepest bills in this industry rarely show up at go-live. They show up at renewal, when your options feel smallest and your staff is trained on a system that would take months to replace.

This is also becoming a compliance issue, not just a pricing one. CMS's interoperability and prior authorization rule requires impacted payers to support a FHIR-based Prior Authorization API covering covered-service information, documentation requirements, and request and response functionality (CMS fact sheet, CMS-0057-F). Starting largely in 2026, impacted payers must generally issue decisions within 72 hours for expedited requests and seven calendar days for standard requests, and denials must state a specific reason regardless of how the request was submitted (CMS, 2024). Beginning in 2027, payers must expand Patient Access APIs to include prior authorization data and stand up Provider Access APIs covering claims, encounter, clinical, and authorization data (CMS interoperability FAQ). If your platform can't exchange data on those terms, you're not just overpaying. You're building a manual workaround on top of a system you already pay a premium for.

What a partnership looks like instead

A software relationship should behave like a partnership, which means the incentives are aligned. Month-to-month terms instead of multi-year lock-in. One platform instead of a base fee plus an a-la-carte menu that grows every year. Pricing you can actually see before you sign. A partner earns the renewal each month because the product is worth it, not because leaving is engineered to hurt.

Judge the relationship by results your team can measure, not by the length of the feature list. Watch inquiry to assessment time, assessment to admission conversion, and abandoned referrals. Watch unsigned note aging, schedule fill rate, and how many staff hours go to manual coordination that the system should have handled. Watch clean claim rate, days in accounts receivable, denial rate, and authorization turnaround. If those numbers improve after implementation, the software is earning its keep. If they don't move, or move the wrong way, no amount of added functionality justifies the bill.

The questions worth asking before you sign

A handful of concrete questions cut through most of the fog:

• What is the all in annual cost, including every module we will actually use, for the next three to five years, including implementation, interfaces, additional users, locations, storage, and support tiers?
• What happens to that number at renewal, and is there a cap on the annual increase?
• Which capabilities are core to the contracted price, and which are add ons that could be unbundled later? Get the answer in writing.
• If we leave, what do we take with us? Ask for machine-readable exports, complete clinical and financial records, documented data schemas, and a defined transition-assistance timeline.
• Does the system support standards-based data exchange for payer authorizations, or will our staff be re-entering the same information twice?

If a vendor cannot or will not answer these plainly, you have learned something important before you have signed anything.

Track the cost per outcome, not the cost per feature

The right benchmark is not how many capabilities a platform has accumulated. It's whether cost per treated patient, cost per completed admission, administrative hours per episode, and revenue lost to preventable delays are getting better or worse. Treat every renewal as a chance to re-run that math, not a formality to sign and forget.

A good partner grows with you and shows its pricing without being asked twice. A lock-in arrangement just sends an invoice, and raises it.