For a growing number of Americans, the path to behavioral health care does not begin in a doctor’s office or with a referral from a friend. It begins with a search box. Before anyone fills out an intake form or sits in a waiting room, they type a question into a search engine, and what generates in the minutes after that query often determines whether they get care at all. As someone who builds consumer-facing tools in this space and spends a great deal of time studying how people phrase these searches, I have come to believe that the search-to-care gap is one of the most under-examined access problems in behavioral health. It is also one of the most fixable.
Insurance Comes First, Not Last
The single most consistent pattern in behavioral health search behavior is that insurance is not an afterthought; it is often in the very first query. People do not typically search “therapist near me” and then work out payment later. They search “therapist near me that takes Blue Cross,” “psychiatrist that accepts Medicaid,” or “does Medicare cover therapy.” In plain terms, people search for the name of the card in their wallet.
Search demand data bears this out at a granular level. Query volume clusters at the intersection of geography and payer: a state name plus an insurance type, sometimes with a condition or care format attached. Someone in Texas on Medicare is not searching for behavioral health care in the abstract; they are searching for the specific, narrow slice of the system that will see them at a price they can manage. Cost consistently ranks among the most cited barriers to behavioral health treatment, and these queries reflect that reality. The affordability question is asked before the clinical one.
This matters for anyone who publishes information about behavioral health services. Most provider websites and health system content are organized the opposite way: by specialty, by clinician, by service line, with insurance buried on a billing page, listed as a wall of carrier logos, or missing entirely. The field is publishing payer-last answers, but consumers are asking payer-first questions.
Where People Get Stuck
If you follow the consumer journey from query to appointment, you will see the same four failure points consistently.
- The directory dead end. Insurance provider directories are the official answer to the payer-first question, and they are notoriously unreliable. In a 2023 secret shopper study by the U.S. Senate Committee on Finance, staff called 120 mental health providers listed as in-network across 12 Medicare Advantage plan directories in six states. A third of the listings had inaccurate or non-working numbers or calls that were never returned, and staff secured an appointment only 18 percent of the time. More than 80 percent of the listed providers were, in the committee’s words, “ghosts.” They were simply unreachable, not accepting new patients, or not actually in network. For a consumer, every ghost listing is not just an inconvenience; it is a discouraging phone call at a moment when discouragement is exactly what they cannot afford.
- The plan-name problem. Consumers do not search using the industry’s taxonomy. A person on a Medicare Advantage plan usually searches the carrier’s brand name, not “Medicare Advantage.” A person with employer coverage searches “does [carrier] cover therapy” without knowing that the answer depends on their specific plan’s network, not the carrier’s logo. Queries containing “Medicare” blend people on Original Medicare, Medicare Advantage, and supplemental plans, which can produce meaningfully different answers. Content and directories written in industry language routinely fail to match how the question is phrased, so the consumer either gets no answer or, worse, the wrong one.
- The cash-pay collision. Search results for insurance-driven queries are frequently dominated by online therapy platforms that operate primarily on a cash-pay or subscription basis. There is nothing wrong with those services, and for many people they are a genuinely useful and affordable option. However, a person who searched “therapy that takes Medicare” and landed on a service that does not accept Medicare often discovers the mismatch only deep into the sign-up process. The result is wasted effort, eroded trust, and another abandoned attempt. Clear, early disclosure of what coverage a service does and does not accept would resolve most of this friction, and the services that disclose it plainly tend to serve consumers better.
- The last mile of intake. Even when a search finds a plausible, in-network, accepting provider, the final step is frequently a phone number and nothing else. Voicemail, unreturned calls, and opaque waitlists are where much of the remaining intent quietly dies. The Senate study’s 18 percent appointment rate captures this last mile as much as it captures directory inaccuracy. Finding a provider and starting care are two different achievements, and the distance between them is longer than most organizations realize.
The Ground Is Shifting Under Search Itself
This picture is now changing quickly, because search itself is changing. According to the 2026 Patient Choice Report from rater8, AI tools such as ChatGPT and Claude are now cited by 36 percent of patients as an influence when choosing a new provider. This is slightly ahead of traditional Google search results at 34 percent and physician referrals at 32 percent. When asked which part of a Google results page they trust most, more patients chose the AI-generated overview than the traditional organic links. For behavioral health organizations, the implication is significant. AI systems synthesize whatever data is available upstream. If directory listings, plan acceptance information, and availability data are inaccurate today, those inaccuracies will continue to mislead consumers using the tools they increasingly trust.
What Behavioral Health Organizations Can Do
None of these failure points require new legislation or new technology to address at the organizational level. A few practical steps go a long way.
Treat your listings as care infrastructure. An inaccurate directory entry functions as a barrier to care just as surely as a waitlist does. Auditing your organization’s listings across the payers you contract with and correcting them with the same urgency you would apply to a clinical error is a meaningful access intervention.
Answer the payer-first question in plain language, on pages you control. Your own website is the one listing no insurer can get wrong on your behalf. A simple, prominent, dated statement of which plans you accept, whether you are taking new patients, and roughly how long a new patient waits matches the question consumers are asking and spares them a phone call that too often goes unanswered.
Meet the query where it lives. People search by state, payer, and situation. Educational content that directly addresses those intersections — what a given coverage type does and does not typically include in your state, and what a realistic path to care looks like — serves consumers far better than generic service-line pages.
Give intake a second door. A web-based intake request, a callback option, or even a monitored email address catches the many people for whom one more unanswered phone call is one too many.
The people typing these searches are, by definition, already motivated to seek help. They have crossed the hardest threshold on their own. The queries themselves are visible to anyone who cares to look, and they tell us with unusual precision where the system loses people. Closing the gap between the search and the session may be one of the most cost-effective access improvements available to the field.
Patrick Garaca is the founder of PsychCare.ai, a free consumer tool that helps people navigate behavioral health care options based on their state, coverage, and preferences. You can reach him at [email protected] or via psychcare.ai.
This was excellent. However, I’ve read one major reason insurers don’t update their provider listings is that it makes them look better having all those provider listings to potential employers AND members shopping for healthcare plans when they do get to choose among several healthcare plans. This is a version of the classic “bait and switch” approach in my humble opinion. Yes, shoppers can confirm their current providers actually take the insurance they’re shopping for, but do they really do that for every clinician they see? Also, if you’re a healthy young adult seeing few providers, you may not even think to confirm/check and be reassured you could surely find someone near home or work from that lengthy list. Food for thought….