How Long Does Therapy Actually Take?

Clinical examples in this article are fictional composites created for educational purposes and do not describe any one client.

It’s one of the first questions people ask, and one of the hardest to answer honestly. The unhelpful answer is “it depends.” The honest version: it depends on specific things, like what you’re working on, what approach you’re getting, and what “done” means for you. Anyone who gives you an exact number before they’ve assessed you is guessing. Here are the real ranges.

How Long Does Therapy Take for Each Concern?

It depends on the problem. In my practice, moderate OCD with ERP usually takes about 12 to 16 weekly sessions. Treat that and the other ranges below as rough guides for structured, goal-focused work, not promises. Every one of them assumes a therapist who knows what they’re doing and is using the right approach for the problem.

OCD (with proper ERP): about 12-16 sessions

Here’s the honest version for ERP (Exposure and Response Prevention), a structured treatment with a clear arc. In my practice, people with moderate OCD usually finish the active part of ERP in about 12 to 16 weekly sessions; it can be shorter or longer. You’ll get an honest estimate after your assessment, not a promise. What moves it most, in my experience: how severe things are, and whether you do the exposures between sessions. It’s like training with a coach at the gym. The plan has a shape, and you can see where you are in it. The critical qualifier: this is ERP specifically, not general talk therapy about OCD. If you’re getting the latter, there’s no real timeline, because the treatment isn’t aimed at the loop. If you’re looking for an OCD-focused therapist, here’s how to find an OCD therapist.

Anxiety disorders: 8-16 sessions

For generalized anxiety, social anxiety, panic disorder, and specific phobias, cognitive-behavioural approaches tend to produce meaningful results within 8-16 sessions. The more specific the fear (one phobia, one clear trigger), the faster the work tends to go. Anxiety that follows you everywhere and attaches to whatever’s available usually takes longer, because you’re working on a pattern, not a single target. It’s like whack-a-mole: knock one worry down and the next one pops up.

Depression: 12-20 sessions

The evidence for structured treatment of depression, particularly behavioural activation and CBT, supports meaningful change within 12-20 sessions for mild to moderate presentations. Severe or recurrent depression may require a longer initial course of treatment, and some people return for maintenance work during high-risk periods. The key distinction: a first episode of depression in an otherwise stable life often responds faster than depression layered on top of years of difficult circumstances.

Couples therapy: 12-24 sessions

Couples work is inherently slower because you're dealing with a system, not a single person. You're also often addressing patterns that have calcified over years. Couples who come in with an acute crisis, like infidelity or a major life disruption, may work faster because the stakes create urgency. Couples who've been grinding in conflict for a decade typically take longer. If OCD is part of the strain, here’s how couples therapy for OCD works.

Single-incident trauma: 8-16 sessions

When the trauma is one clear, bounded event, rather than chronic or developmental trauma, evidence-based treatments like EMDR and prolonged exposure tend to work relatively efficiently, often within 8-16 sessions. Complex trauma, childhood trauma, and trauma that's woven into someone's sense of self typically require more time and a different framing.

Life stress and transitions: 6-16 sessions

For people going through a difficult period, like a career change, a breakup, a loss or a move, shorter-term work often makes the most sense. Six to sixteen sessions is typical for focused work around a specific life stress. The goal isn't deep structural change; it's developing clarity, coping, and forward motion. Some people find enormous value in 8-10 sessions and then stop. Others use that initial work as a foundation and return later.

What Does "Done" Actually Mean?

Done means you’ve reached the goals you named at the start, so how long therapy takes is really a question about endpoints. Have that conversation with your therapist, early.

Some therapists work toward specific, measurable goals: running your exposures on your own, getting back the hours the rituals were taking, going to the grocery store without the compulsion. When those goals are met, there’s a natural conversation about whether more work is needed or whether it’s time to end. This is my model. Goals are first downs, not touchdowns: you can see the chains moving. I believe in working toward something defined, not “continuing as long as it’s useful” without checking whether it still is.

Other therapists approach therapy as longer-term relational work: building self-understanding, resilience, and the capacity to process difficulty. This is also legitimate. It just has different logic: the endpoint isn't a symptom score but a quality of self-knowledge and functioning that's harder to measure but real.

Some clients use therapy more like insurance. They do intensive work, stabilize, and end formal treatment, but value having an ongoing relationship to call on when significant events happen. A parent dies. A marriage falls apart. A health crisis. They don't need weekly sessions, but they want access to someone who knows them. This is a legitimate use of therapy too, as long as both parties are clear that's what's happening.

The question to ask your therapist early: "When does therapy end in your view?" The answer tells you a lot about how they think about the work, and whether that model matches what you're actually looking for.

Why Does Some Therapy Go On Too Long?

Mostly because there are no milestones and nobody stops to check. That part isn’t said plainly very often.

Some therapy goes on for years without clear goals. Not because the client needs it, but because nobody’s asked whether it’s still necessary. There are a few dynamics that drive this.

No milestone structure

If there’s no framework with milestones (here’s what progress looks like, here’s how we’ll know we got there), there’s no natural checkpoint. Sessions continue because sessions are what you do on Tuesdays. This isn't necessarily malicious, but it isn't a plan.

The therapy relationship itself becomes the goal

Some clients genuinely need the relationship, particularly people who grew up without stable, reliable attachment figures. The therapy relationship itself is the corrective experience. This is real and not nothing. But it's worth being clear about: are you continuing because you're still doing work, or because the relationship feels important to maintain? Both can be true simultaneously, but they're worth distinguishing.

Insight-oriented work has no natural endpoint

Open-ended, insight-oriented therapy, the kind descended from psychoanalytic traditions, doesn’t have built-in endpoints. You’re always becoming more self-aware. There’s always more to understand. For some people, that open door is the point, and insight versus action is a real choice between two types of therapy. But it means you need to impose your own structure: "What am I working toward, and how will I know when I've gotten there?"

Nobody's asked the obvious question

Sometimes therapy goes on too long simply because nobody asks: "Are we getting somewhere?" If you've been in therapy for more than six months and can't clearly articulate what's changed or what you're working toward, that conversation is overdue. You can ask it directly, and a good therapist will welcome it. “I want to check in on whether we’re making progress. Can we talk about that?” is a completely reasonable thing to say. Here's how to know if therapy is actually working.

What Does the Average Look Like?

At our clinic, the average client stays 10-12 sessions, typically spread over 3-4 months. That's not a cap. It's what meaningful change often takes for focused, goal-oriented work. Some clients need more. Some need less. But it's a useful baseline for planning.

It's worth thinking about this as a total investment rather than an open-ended commitment. Ten to twelve sessions at $225 to $250 per session is $2,250 to $3,000. That's not nothing. But it’s also not the endless drain some people picture when they think about starting therapy. It’s a course of treatment with an end in view.

In my experience, the thing that stretches timelines most is starting with the wrong approach. With OCD, therapy that keeps analysing the meaning of the thought, or reassuring you, can feed the cycle instead of breaking it. Depression treated with purely supportive listening may feel validating but may not produce lasting change. The approach matters as much as the duration. If you want to know what the right approach looks like for what you're dealing with, the consultation is the place to start. Getting the method right from the beginning is the most efficient thing you can do.

The other thing worth naming: ending too early isn’t always efficient. Stopping at session 6 when the real work starts at session 8 can mean starting over. With ERP, the last stretch is where you practise running the exposures without me in the room. It’s just reps. A few extra sessions to lock in what you’ve learned are usually worth it.