Does Couples Therapy Work? What the Research Says

Yes, for many couples. Across dozens of studies, couples who go through therapy report clearly higher relationship satisfaction than couples who wait, and many keep those gains for years. It does not work for everyone. In long-term trials, roughly half of couples reach lasting, meaningful improvement, and results depend on commitment, timing and fit with the therapist.

What does the research say overall?

The short answer from several decades of studies is that couples therapy helps most couples who complete it, at least in the short term.

The largest recent summary is a 2020 meta-analysis in the Journal of Consulting and Clinical Psychology. It pooled 58 studies covering 2,092 couples. Couples in therapy showed a large improvement in relationship satisfaction from start to finish. Couples placed on a waiting list did not improve significantly over the same period. In other words, distressed relationships did not tend to fix themselves by waiting. The same analysis found gains in communication, emotional intimacy and how partners behaved toward each other.

A 2022 review of the field in Family Process puts it this way: the average person who receives couple therapy ends up better off than 70 to 80 percent of people who do not receive it. That is an average across many couples. It is not a promise about any one relationship.

Couples therapy can also help when one partner is struggling individually. A Cochrane review of 14 studies on couple therapy for depression found signs that it was about as effective as individual therapy for depressive symptoms and better at reducing relationship distress, although the authors rated the quality of that evidence as low or very low.

What does “works” mean in these studies?

This is where headlines mislead. Most studies measure relationship satisfaction with a questionnaire before and after therapy. “It worked” usually means the average score went up by a meaningful amount.

That is different from saying every couple ended up happy. Researchers use a stricter test called clinically significant change, which asks whether a couple moved from the distressed range into the range of satisfied couples. A review of 20 meta-analyses of marriage and family therapy concluded that therapy produces clinically significant results in about 40 to 50 percent of those treated.

So a realistic reading is this: most couples improve, and somewhere near half improve enough to no longer look distressed on standard measures. A single “success rate” quoted without a source and a definition deserves skepticism.

It is also worth knowing that staying together is not the only good outcome. Some couples use therapy to reach a clear, respectful decision to separate. Studies that count only intact relationships as success miss that.

Do the gains last?

For many couples, yes. For some, the benefit fades.

One of the longest follow-ups is a randomized trial of 134 chronically and seriously distressed married couples who received about eight months of behavioral couple therapy and were then tracked for five years. At the five-year point, about half of the couples (50 percent in one treatment group and 46 percent in the other) still showed clinically significant improvement compared with where they started. Roughly a quarter had separated or divorced.

Other research is less encouraging about durability. A meta-analysis limited to randomized controlled trials found medium-sized gains at the end of therapy and smaller gains six months later. For behavioral couple therapy, the one approach with 12-month data, the advantage over comparison groups was no longer evident at a year. The authors also warned that publication bias may inflate the picture.

The practical lesson: skills fade if you stop using them. Many couples benefit from tapering sessions gradually and coming back for a check-in when old patterns return.

What are the limits of the evidence?

Good research is honest about what it has not shown. Four limits matter if you are deciding whether to try therapy.

• Who was studied. The 2020 meta-analysis notes that its sample was made up exclusively of different-sex couples. Much less is known from controlled trials about same-sex couples, unmarried couples and couples from many cultural backgrounds.
• Trials are not everyday practice. The Family Process review notes that results in ordinary clinical settings appear somewhat lower than in closely monitored trials.
• Small studies. A 2026 systematic review in the Journal of Marital and Family Therapy applied a strict set of criteria to 40 studies and found that no couple therapy model met the standard for strong empirical support, mainly because of small samples and a lack of independent replication. Seven models, including emotionally focused and Gottman approaches, showed modest support.
• Short follow-up. Few studies follow couples for more than a year or two.

None of this means therapy is ineffective. It means the honest claim is “helps many couples” and not “works for nearly everyone.”

Does the method or the therapist matter more?

Differences between well-studied methods are generally small. The review of 20 meta-analyses reached that conclusion, and the later meta-analysis of randomized trials found no significant difference between emotionally focused and behavioral couple therapy. If you want a side-by-side look at three well-known approaches, see Gottman vs. EFT vs. Imago.

What shows up more consistently is the working relationship with the therapist. A meta-analysis of 48 studies on the therapeutic alliance in couple and family therapy found that a stronger alliance was linked to better outcomes. It also found that a “split” alliance, where one partner feels allied with the therapist and the other does not, was linked to poorer outcomes. Both of you need to feel the therapist is fair.

What makes it more likely to work for you?

Research on who benefits points to a few things you have some control over.

• Commitment. In the five-year follow-up described above, couples with higher commitment at the start were less likely to divorce or separate.
• Not writing yourselves off. The 2020 meta-analysis found that couples who started out more distressed tended to make larger gains. Being in a bad place does not rule out improvement. More on that in Is it too late for couples therapy?
• Both partners feeling heard. If one of you feels ganged up on, say so in the session. A good therapist will want to know.
• Practice between sessions. A session is a small part of your week. What you try at home in the days between is where new habits form.
• Enough time. Research programs typically run for months, not two or three visits. See how long couples therapy takes for realistic timelines.

When is it worth getting professional help?

It is worth a conversation with a therapist if the same argument keeps repeating, if you avoid each other to keep the peace, if trust has been broken, or if one of you has started to think about leaving. You do not need to wait for a crisis. The waiting-list findings above suggest that serious distress rarely resolves by itself.

There is one important exception. If there is physical violence, threats, coercion, or you are afraid of your partner, couples therapy is not the right first step. Contact the National Domestic Violence Hotline at 1-800-799-7233 for confidential support. Other situations where joint sessions are not advised are covered in when couples therapy is not recommended.

At spiral2grow, couples counseling with Moshe Ratson, LMFT, uses 45-minute sessions that are usually weekly, and the counseling process starts with clarifying what each of you wants to change.

Talk it through before you decide

If you are weighing whether therapy is worth trying, a short conversation can help. Moshe Ratson, LMFT, offers a free phone consultation and sees couples in Midtown Manhattan and Online by Video. You can request a consultation through the contact page and ask your questions before booking anything.

Common questions

Sources

• Meta-analysis of couple therapy: Effects across outcomes, designs, timeframes, and other moderators (Roddy et al., 2020), Journal of Consulting and Clinical Psychology, via PubMed
• Couple therapy in the 2020s: Current status and emerging developments (Lebow & Snyder, 2022), Family Process, via PubMed Central
• Couple therapy for depression (Barbato, D’Avanzo & Parabiaghi, 2018), Cochrane Database of Systematic Reviews, via PubMed
• Meta-analysis of MFT interventions (Shadish & Baldwin, 2003), Journal of Marital and Family Therapy, via PubMed
• Marital status and satisfaction five years following a randomized clinical trial comparing traditional versus integrative behavioral couple therapy (Christensen et al., 2010), Journal of Consulting and Clinical Psychology, via PubMed
• The Efficacy of Emotionally Focused Couples Therapy and Behavioral Couples Therapy: A Meta-Analysis (Rathgeber et al., 2019), Journal of Marital and Family Therapy, via PubMed
• Empirically Supported Couple Therapy Models for Couple Relationship Distress: A Replication and Update (Jackson et al., 2026), Journal of Marital and Family Therapy, via PubMed Central
• Meta-analysis of the alliance-outcome relation in couple and family therapy (Friedlander et al., 2018), Psychotherapy, via PubMed
• Prediction of treatment response at 5-year follow-up in a randomized clinical trial of behaviorally based couple therapies (Baucom et al., 2015), Journal of Consulting and Clinical Psychology, via PubMed
• National Domestic Violence Hotline, National Domestic Violence Hotline

This article is part of the Couples Counseling Guide. It is general information, not a substitute for care from a licensed professional who knows your situation.

This article is for general education. It is not a substitute for professional advice, diagnosis or treatment.