Is Therapy Worth the Investment? The ROI of Healing

STARTING THERAPY • April 6, 2026

The Spreadsheet Before the First Session

It’s 9:40 on a Tuesday night, and Sofia is sitting cross-legged on her kitchen floor with her laptop balanced on a cutting board, because the dishwasher is running and the kitchen table is covered in her daughter’s school project. She’s 41, a director of product at a mid-size fintech company, and she has three browser tabs open. One is a therapist directory. One is a spreadsheet she built an hour ago, columns for session cost, estimated number of sessions, insurance reimbursement percentage, and a column she has labeled, with something like embarrassment, “expected benefit.” The third tab is just her email, open to a message from her sister that says, in full, “have you called anyone yet.”

She has not called anyone yet.

“I keep trying to figure out if this is a good use of money,” she tells me, weeks later, once she has finally called someone. “Which sounds insane when I say it out loud. I spend more than a therapy session costs on a mediocre dinner out and I don’t build a spreadsheet for that. But somehow this feels like it needs to prove itself first. Like I need to know it’s going to work before I’m allowed to start.”

I felt something familiar sitting across from Sofia that first session, something I have felt with a striking number of driven, capable women over the years. Not confusion about her hesitation. Recognition. The spreadsheet was not really about money. The spreadsheet was about control, about needing a guarantee before she was willing to be vulnerable, about a nervous system that had learned, probably decades before that Tuesday night on the kitchen floor, that uncertainty was dangerous and competence was the only safe way through.

Here’s the thing I want to say clearly, before we go any further: there is no spreadsheet that can answer this question. Not because the question doesn’t matter, but because “is therapy worth the investment” is not actually a financial question, even though it gets asked like one. This piece is going to walk through the research on why therapy helps some people and not others, what “return on investment” can and cannot mean here, and how insurance, access, and systemic barriers shape who even gets to consider this question in the first place. It’s not going to hand you a number. I don’t think a number exists, and I think it’s worth being honest about that upfront.

What “ROI of Healing” Actually Means (And Doesn’t)

Let’s define our terms, because “ROI of healing” is a phrase I’m using deliberately as a metaphor, and I want to be precise about what it is and is not.

What I can tell you, based on the research and on more than fifteen years of clinical work, is something more modest and, I’d argue, more honest: therapy helps many people, to varying degrees, depending on the person, the condition, the clinician, the fit between them, the modality, the duration of treatment, and the person’s access to consistent care. That is not a marketing sentence. It is a hedge, and it is a hedge because the research itself is hedged. A 2022 umbrella review in World Psychiatry, synthesizing evidence from over one hundred meta-analyses covering more than 3,700 randomized trials and over 650,000 patients, found small to moderate average benefits across psychotherapies and pharmacotherapies for adult mental health conditions, with meaningful variation across conditions and treatments (Leichsenring et al. 2022). Small to moderate, on average, with real variation. That is what rigorous research says. It is not a story about guaranteed transformation, and I don’t think it needs to be in order to be worth taking seriously.

So if “ROI of healing” cannot mean a calculable financial return, what can it mean? I think of it as a metaphor for something closer to this: whether the investment of time, money, vulnerability, and emotional effort feels, in retrospect, aligned with what actually mattered to you. That is a subjective, values-based question. It is not a question a spreadsheet can answer, and treating it like one, the way so many driven women approach overworking and self-optimization more broadly, is often itself part of the pattern that brought someone into therapy to begin with.

What the Research Says About Why Therapy Helps

Before we get into the research, it’s worth naming plainly that relational trauma therapy specifically is one context where this question comes up constantly, since the presenting concern so often looks like a work problem or an anxiety problem on the surface, and turns out to have deeper roots. I recently found myself returning to a 2015 paper by Bruce Wampold, PhD, psychologist and one of the field’s most cited researchers on psychotherapy outcomes, and I haven’t been able to stop thinking about it since. In How Important Are the Common Factors in Psychotherapy?, published in World Psychiatry, Wampold synthesizes decades of outcome research and lands on a finding that runs counter to how most people, including many clinicians, talk about therapy (Wampold 2015). The specific modality, EMDR versus CBT versus psychodynamic versus somatic work, accounts for a relatively small portion of the variance in outcomes. What accounts for more of it is what researchers call the common factors: the strength of the therapeutic alliance, the client’s expectations, the therapist’s skill in adapting to the individual person in front of them, and the client’s own engagement in the process.

Here is where I want to bring in the body, because driven women often experience their struggle as a thinking problem long after it has become a nervous system problem. Bessel van der Kolk, MD, psychiatrist and trauma researcher, has spent decades studying what happens in the brain and body under chronic or acute stress. In a 2006 paper on the clinical implications of neuroscience research in PTSD, he describes how traumatic reminders activate brain regions involved in intense emotion while decreasing activation in regions responsible for integrating sensory input, regulating physiological arousal, and putting experience into words (van der Kolk 2006). Which is the clinical way of saying something I hear from clients constantly, in some version of these words: “I understand exactly why I do this. I’ve read the books. I still can’t stop doing it.”

Understanding is Layer One. It is not nothing, but it is not the same as change. What van der Kolk’s research points to is the reason talk alone sometimes plateaus: if a pattern is held partly in the nervous system rather than only in narrative memory, then talking about the pattern accurately can coexist with the body still reacting to it as though the danger were current. This is one reason effective treatment for trauma-related presentations often includes some attention to physiological regulation, not because insight doesn’t matter, but because it is frequently necessary and not always sufficient on its own.

I want to be precise about what this research does and does not support. It does not mean any single modality, EMDR or somatic work or otherwise, is guaranteed to work for any given person, or that it “heals the root cause” for everyone who tries it. It means researchers have identified plausible mechanisms for why body-inclusive approaches sometimes help people who feel stuck in insight-only work, and that this varies by person, condition, and clinician fit. If you want to read more about how this plays out specifically around relational trauma, I’ve written at length about betrayal trauma and the nervous system elsewhere, and the same caveats about individual variation apply there too.

There’s a related piece of this that I think gets lost in ROI framing entirely, which is the question of duration. The research doesn’t support a single answer to “how many sessions until it works,” because the honest answer is that it depends on what’s being treated. A specific phobia responds differently than complex developmental trauma. A single-issue presentation responds differently than a longstanding relational pattern with roots in childhood. Some clients feel meaningfully different in eight to twelve sessions. Others are doing foundational work for a year or more before they’d describe a shift as durable. Neither timeline is a sign of success or failure. It’s a sign of what was actually being addressed, which is one more reason a fixed ROI calculation doesn’t hold up: the input itself, the length and intensity of treatment, isn’t standardized, and shouldn’t be.

What I notice clinically is that driven women often want the phobia-style timeline, the fast, contained, clearly bounded course of treatment, even when what they’re actually dealing with is closer to the longstanding relational pattern. That mismatch between expectation and reality is itself worth naming early, ideally in the first or second session, so that six months in, a client isn’t quietly deciding the work “isn’t working” simply because it doesn’t resolve on the timeline she’d unconsciously budgeted for it.

How This Shows Up in Driven Women’s Decision-Making

Sofia called the therapist from the directory tab, eventually. Not immediately. It took another eleven days, three more spreadsheet revisions, and, she told me later, one very specific moment: she was in a client meeting, presenting a roadmap she had built with total confidence, and she felt her chest tighten so hard she had to excuse herself and stand in a stairwell for four minutes before she could go back in and finish the meeting like nothing had happened.

“I closed the deal,” she said. “Nobody in that room knew what had just happened to me. I got in my car afterward and I just sat there. I thought, I cannot keep doing the thing where I perform being fine in front of everyone and then fall apart alone in a parking garage. That’s not sustainable. That’s not a life.”

Sitting with Sofia as she described that stairwell, I felt the particular ache I feel often with women who have built genuinely impressive lives on top of a nervous system that never quite got the message that the danger had passed. The performance was not fake. She really had closed the deal. She was really that competent. And she was also, underneath the competence, running on a kind of vigilance that had likely been installed long before her first job, in a family system where being fine, being helpful, being the one who didn’t need anything, kept things calmer for everyone else.

What I’ve come to think of as the parking-garage moment is something I see across driven women almost weekly: the instant where the cost of continuing to perform wellness becomes more exhausting than the perceived risk of finally addressing what’s underneath it. It rarely arrives as a calm, rational decision. It usually arrives as a body that simply will not cooperate with the performance anymore. If any of this sounds familiar, particularly the pattern of being fine everywhere except alone, I’ve written specifically about what it feels like when your life looks good from the outside and empty from the inside, and it’s a pattern I want to be careful not to oversimplify here.

Mariah, 40, a partner at a regional law firm, came to this decision from a different angle entirely. Where Sofia needed a crisis to override her spreadsheet, Mariah came in having already made peace with the fact that she couldn’t know in advance whether it would “work.” What she couldn’t tolerate was something else. “I’ve built my entire career on being the person who solves problems,” she told me in our first session. “And I have a problem I cannot solve by myself. That’s the part that’s hard to sit with. Not the money. The admitting I can’t strategize my way out of this one.”

What strikes me about both of these women, and about so many driven, ambitious clients I’ve worked with over the years, is how rarely the actual hesitation is about the number on the invoice. It’s almost always about something underneath the number: a fear of what they’ll find if they stop managing everything for fifty minutes, a belief that needing help is itself evidence of failure, or a lifetime of being the one who holds things together for everyone else and has no template for being the one who receives care instead. I’ve written before about how the very engine that produces success in driven women is often the same engine that’s quietly costing them, and the decision to finally get support often requires confronting that engine directly, which is uncomfortable in a way no spreadsheet prepares you for.

Sofia’s second return to this theme came about ten weeks in, once the acute crisis had settled and the harder, slower work had begun. “I keep waiting for the moment where I can prove to myself this was the right call,” she said. “And I don’t think that moment is coming. I think I just have to decide it was right without the proof.” I told her that was, in my experience, close to the most honest place anyone gets to on this question. Not certainty. A decision to proceed without it.

Therapy, Coaching, and the Other Paths Available to You

Part of what makes this decision harder than it needs to be is that “getting support” is often talked about as one undifferentiated category, when it’s really several distinct things, each suited to different situations.

There are other legitimate paths too, and I want to name them because the conversation about “is therapy worth it” sometimes implies therapy is the only serious option, which isn’t accurate. Consultation with a psychiatrist or primary care physician matters when a physical health condition or medication question is part of the picture. Peer support, whether structured groups or informal community, offers something therapy cannot: the specific relief of being among people who have lived the same thing. Structural or workplace changes, such as adjusting a role, a schedule, or a management relationship, sometimes address the actual source of distress more directly than any individual-level intervention could. And self-directed education, reading, journaling, workshops, can be meaningful groundwork, even if it isn’t a substitute for clinical treatment when clinical symptoms are present.

I think about that Mary Oliver line often in this context, not as a prompt toward a bigger, louder life, but as a genuine question worth sitting with before defaulting to any one path. What do you actually want your capacity, your attention, your one life, to be used for? That question, more than any spreadsheet, is usually the more honest starting point.

Mariah eventually added an executive coach to her therapy, about eight months in, once the therapeutic work had stabilized some of what was driving her decision paralysis at work. “I needed the therapy first,” she said. “The coaching wasn’t going to touch what was actually happening. But once that shifted some, the coaching helped me figure out what I wanted to do with a version of myself that wasn’t constantly bracing for impact.” If this distinction is useful to you, I’ve written more about how early-life roles shape adult decision-making capacity, which is often the deeper layer underneath “should I get a coach or a therapist.”

For some driven women, the more useful comparison isn’t therapy versus coaching in the abstract, but a specific professional context that’s actively shaping the decision. If your hesitation is tangled up with a demanding professional track that seems to reward exactly the overfunctioning you’re trying to examine in therapy, I’ve written specifically about what happens when a high-stakes career track gets into your nervous system, and how therapy and coaching sometimes need to work in tandem rather than in sequence. Similarly, if the calculation you’re running involves whether stepping back from an intense role would even be possible, the dynamics I’ve described around being promoted into an unsupported, high-risk role are often part of what makes the decision to invest in support feel urgent rather than optional.

Both/And: It Can Be Worth It and Still Not Guarantee Anything

Here is the both/and I want to hold carefully, because I think oversimplifying it in either direction does a disservice to the reader trying to make a real decision with real constraints.

Both: the research on therapeutic alliance and common factors gives genuine reason to believe that good-fit therapy, with a clinician you trust, meaningfully helps many people reduce distress and change longstanding patterns (Wampold 2015). And: therapy does not guarantee a specific outcome for any individual person. It does not promise a promotion, a saved marriage, a resolved chronic illness, or a fixed number of avoided sick days. Effectiveness varies by condition, by person, by clinician, by the quality of fit between client and clinician, by duration of treatment, and by modality. Some people improve substantially. Some improve modestly. Some don’t find the right fit on the first, second, or third try. All of that is real and none of it invalidates the value of trying.

Sofia asked me directly, around session six, whether I could tell her it was going to work. I told her the truth: I couldn’t promise that, and anyone who promises it isn’t being fully honest with her. What I could say was that the alliance between us felt solid so far, that she was engaging with real honesty, and that those two things are among the better predictors we have, even though “better predictor” is a long way from “guarantee.” She sat with that for a moment. “That’s more honest than I expected,” she said. “I think I needed someone to just say that instead of selling me on it.”

This same both/and shows up in how driven women relate to rest and recovery more broadly, not just to therapy specifically. I’ve written before about why rest itself can feel unsafe for people whose nervous systems learned that stopping was dangerous, and the same resistance often shows up around starting therapy: it requires stopping the performance long enough to be honest about what’s underneath it. Both: the cost of a therapy hour is real, and for many people it is a genuine financial constraint that deserves to be taken seriously rather than moralized about. And: the cost of continuing to live inside an unaddressed pattern is also real, even though it resists being priced. Neither cost cancels the other out or proves the other one is the “right” choice. This is not a case where naming the systemic and financial constraints makes the emotional stakes disappear, or where naming the emotional stakes makes the financial constraints disappear. Both are true at once, and a genuine decision has to hold both.

Mariah’s version of this both/and showed up around month four. She’d started noticing real shifts, sleeping better, arguing less reflexively with her husband, catching herself mid-pattern instead of only after the fact. And she also had a week where an old dynamic with her mother resurfaced so intensely that she wondered, out loud in session, whether therapy had made things worse rather than better. “Both of those are true,” I told her. “You are genuinely doing better on the whole. And this week genuinely felt worse. Healing isn’t a straight line upward, and a bad week doesn’t erase the trend, any more than one good month proves the trend is permanent.” She sat with that for a long moment. “I wanted it to be simpler,” she said. “I wanted a chart going up and to the right.” I understood that completely. Most driven women do. It’s rarely how the actual process moves.

The Systemic Lens: Who Gets to Ask This Question at All

I want to be honest about something the ROI framing tends to obscure: not everyone gets to ask “is this worth the investment” as a matter of preference. For a lot of people, the more accurate question is “can I access this at all,” and that is a structural question, not a personal one.

Insurance coverage for mental health care varies enormously by plan, by state, and by whether a provider is in-network, and that variation shapes who can even get to the spreadsheet stage. Research on racial and ethnic disparities in mental health care access has documented, consistently, that racial and ethnic minority groups in the United States have less access to mental health services than white patients, are less likely to receive needed care, and more often receive lower-quality care when they do, with disparities rooted in unequal insurance coverage, provider availability, and, in some cases, bias in clinical encounters (Cook et al., Health Affairs). That is not a footnote. It’s a structural fact that shapes the entire landscape this article is being written into.

Geography matters too. Licensure is state-specific, which means someone in a rural area may have a much smaller pool of local, in-person providers, even before considering insurance. Telehealth has expanded access substantially for many people, but it does not solve for everyone, particularly those without reliable private space or internet access. Caregiving load, whether for children, aging parents, or both, shapes who has an uninterrupted fifty minutes available on a weekday. Waitlists for in-network providers in many areas run months long, which means the decision to start therapy and the ability to start therapy are frequently two separate timelines. And cultural fit matters enormously: finding a clinician who understands a client’s specific cultural, religious, or identity context is not a luxury preference, it is often central to whether therapy is safe and useful at all.

None of this means the financial question is unimportant. It means the financial question sits inside a bigger set of access questions that a purely personal-responsibility framing, the kind implied by “just invest in yourself,” tends to flatten. If cost or access is the barrier for you right now, sliding-scale clinics, community mental health centers, employee assistance programs, and university training clinics are worth exploring, and a licensed provider or your insurer’s member services line can usually help you understand what’s actually available given your specific plan and location, which varies enough that I’m not going to generalize about it here. If workplace stress is a significant driver of what you’re dealing with, it’s also worth asking honestly whether the more effective intervention is individual treatment, a structural change at work, or some combination, since some of what looks like a personal breaking point is actually a structural one, particularly in high-pressure professional tracks.

Sofia, for what it’s worth, had a PPO plan with decent out-of-network reimbursement, and she still almost didn’t call, not because of the money exactly, but because of what asking for help represented to her. I want to hold both facts without collapsing them: her access was real and relatively good, and her hesitation was still real. Access and readiness are not the same barrier, even when they show up tangled together.

She said something to me around month five that I think about often. “I had every resource I needed to do this sooner,” she said. “Good insurance, a flexible enough job, a sister who kept asking if I’d called anyone. The only thing missing was permission I wasn’t giving myself. And I know that’s not true for everyone. I know some people are dealing with the actual resource piece, not just the permission piece. I got lucky on the resource piece and I still almost didn’t use it.” That distinction, between lacking access and lacking permission, is one I try to help clients hold clearly, because collapsing the two either lets structurally advantaged people off the hook for their own avoidance, or unfairly implies that people without access simply haven’t given themselves permission yet. Both dynamics are real, and they are not the same dynamic.

How to Think About the Decision Without a Spreadsheet

If you’re circling this decision the way Sofia was, sitting on a kitchen floor with tabs open and a column labeled “expected benefit,” here is what I’d offer instead of a formula.

Start with what you can actually verify, rather than what you can guess. Call your insurer’s member services line, or check your plan’s online portal, to find out what your specific mental health benefit actually covers, including in-network versus out-of-network reimbursement, session limits, and copay amounts, since this varies by plan and I can’t generalize it here. If you have an HSA or FSA, confirm with your plan administrator whether therapy and related expenses qualify, since eligibility and documentation requirements vary by plan and by account type. If you have questions about whether therapy expenses are deductible or how they interact with your specific tax situation, that’s a conversation for a qualified tax professional, not something I can answer generally in an article. None of that requires you to know in advance whether therapy will “work.” It just requires you to know what’s actually true about your specific coverage before you decide anything else.

Then, separately, get honest about what you’re actually weighing. Not “will this pay off” in some measurable sense, but something closer to: what does it cost me to keep living exactly as I am right now, and is that a cost I’m willing to keep paying indefinitely? That’s not a rhetorical trick to talk you into therapy. Plenty of people weigh that honestly and decide the timing isn’t right, or that a different form of support fits better right now, and that’s a legitimate answer too.

Sofia started therapy nine weeks after that first spreadsheet tab opened. Mariah started five months before she added coaching. Neither of their timelines is the “correct” one. What I’ve come to think of as the readiness gap, the space between knowing something intellectually and being willing to act on it, closes at different speeds for different people, and it is not a character flaw when it takes longer than you’d like. If you’re in that gap right now, still building the spreadsheet, still not quite calling, that’s not evidence you’re not serious about this. It might just be where you are.

I asked Sofia recently, close to a year into our work together, whether she’d go back and tell her kitchen-floor self anything different. She thought about it for a while before answering. “I’d tell her the spreadsheet was never going to give her the answer,” she said. “I’d tell her to call sooner, not because I can prove it ‘worked’ in some measurable way, but because I’m tired of pretending I can’t tell the difference between how I felt then and how I feel now. I just can’t put a number on the difference. And I’ve stopped needing to.” That’s about as close to an answer as I can offer you, too. Not a return on investment. A woman, a year later, who no longer needs the spreadsheet to know something changed.

If safety is an active concern right now, if you’re having thoughts of harming yourself or someone else, or you’re in a crisis that can’t wait for a scheduled appointment, please contact the 988 Suicide and Crisis Lifeline by calling or texting 988, or go to your nearest emergency room. This article is educational content and is not a substitute for immediate clinical or emergency care.

Frequently Asked Questions

Related Reading

1. Wampold, Bruce E. “How Important Are the Common Factors in Psychotherapy? An Update.” World Psychiatry 14, no. 3 (2015): 270-277. https://pubmed.ncbi.nlm.nih.gov/26407772/
2. van der Kolk, Bessel A. “Clinical Implications of Neuroscience Research in PTSD.” Annals of the New York Academy of Sciences 1071 (2006): 277-293. https://pubmed.ncbi.nlm.nih.gov/16891578/
3. Leichsenring, Falk, et al. “The Efficacy of Psychotherapies and Pharmacotherapies for Mental Disorders in Adults: An Umbrella Review and Meta-Analytic Evaluation of Recent Meta-Analyses.” World Psychiatry 21, no. 1 (2022): 133-145. https://pmc.ncbi.nlm.nih.gov/articles/PMC8751557/
4. Greenman, Paul S., and Susan M. Johnson. “Emotionally Focused Therapy: Attachment, Connection, and Health.” Current Opinion in Psychology 43 (2022): 146-150. https://pubmed.ncbi.nlm.nih.gov/34375935/
5. Cook, Benjamin Lê, et al. “Assessing Racial/Ethnic Disparities in Treatment Across Episodes of Mental Health Care.” Health Services Research, via Health Affairs/PMC. https://pmc.ncbi.nlm.nih.gov/articles/PMC3928067/

Medical Disclaimer