Couples therapy — when it works, when it doesn't

Across multiple randomized controlled trials, EFT shows roughly seventy to seventy-five percent of couples moving from distressed to non-distressed by termination, with about fifty percent maintaining gains at two years. IBCT shows similar effect sizes in the Christensen and Jacobson research, with somewhat better long-term stability for severely distressed couples. The Gottman Method, less heavily studied in RCT form, shows comparable effects in its outcome literature. These are real numbers. They are also limited: studies select for couples without active affairs, severe IPV, or major untreated individual pathology, which describes a meaningful subset of the couples who actually walk into clinical offices. In real-world conditions, success rates are lower, perhaps fifty to sixty percent, still substantial but not the "save your marriage" certainty marketing implies.

A therapist trained in individual work has been taught to align with the client, build rapport with the client, advocate for the client's needs. Drop that therapist into a couples session and they will, often unconsciously, pick one partner to align with — usually the more articulate or more distressed one — and the other partner will, correctly, feel ganged up on. Within three sessions, the unaligned partner is gone, and the therapy has accelerated the bond's collapse. This is not a hypothetical; surveys of couples who have done multiple rounds of therapy show a recurring pattern of "the therapist took her side" or "took his side." A couples-trained therapist is specifically taught to hold the bond as the client, which is a different and harder skill.

The three models with substantial outcome data are EFT (Sue Johnson and colleagues), IBCT (Andrew Christensen and Neil Jacobson), and the Gottman Method (John and Julie Gottman). Each has a distinct theory of change. EFT works through attachment, surfacing the underlying needs beneath conflict cycles. IBCT works through emotional acceptance of differences that cannot be changed and behavior change where it can. The Gottman Method works through specific skills — softened start-up, repair attempts, dreams within conflict — derived from observational research on stable couples. A competent couples therapist usually has primary training in one and working knowledge of the others. Couples shopping for a therapist can reasonably ask which model the therapist primarily uses; an answer of "I'm eclectic" is a yellow flag.

The mean lag between onset of significant relational distress and first couples therapy session is approximately six years, per Doherty and Gottman. The reason is not mystery: it is shame, hope that it will pass, and the fantasy that "we can fix it ourselves." By the time couples arrive, the bond is in chronic dysregulation, contempt is often established, and at least one partner is frequently in a phase of private grief or planning. Couples who treat therapy as preventive — six sessions in year two, a tune-up in year five — get fundamentally different outcomes than couples who arrive at year fifteen with one foot out the door. The cultural framing of therapy as a last resort is itself a major cause of therapy's lower-than-possible success rate.

An active or recently active affair changes everything. The couples frame assumes two partners both inside the bond, however imperfectly. An ongoing affair means one partner is operating with materially different facts than the other, and standard couples interventions become reinforcement of deception. Competent couples therapists screen for affairs, often individually in early sessions, and will refuse couples work until the affair is either fully ended and disclosed or addressed in a different frame (discernment counseling, individual work first). Couples who enter therapy without disclosing an ongoing affair usually waste months and accelerate the damage. The discoverable affair is one of the major sources of "therapy made it worse" stories.

Bill Doherty's discernment counseling protocol is designed for couples where one partner is leaning out and the other is leaning in. Standard couples therapy assumes both partners want to repair; if one is ambivalent, standard models often fail. Discernment counseling, typically five sessions or fewer, helps the couple decide between three paths: status quo, separation/divorce, or a serious six-month commitment to repair work. It is a triage, not a therapy. Many couples who would have wasted a year in couples therapy under false pretenses get a clearer answer in five sessions of discernment work. It is underused because therapists are not trained in it and couples do not know it exists.

A small but real minority of couples leave therapy worse than they entered. The mechanisms are predictable: a therapist who surfaces material the couple cannot regulate around (deep grief, unhealed betrayal, unmet childhood needs) without the skill to contain it; a therapist who unconsciously sides with one partner; a model mismatched to the couple's primary issue. The risk of harm is higher with untrained therapists and with couples already in severe distress. This is not an argument against therapy; it is an argument for choosing carefully. Couples should feel allowed to leave a therapist after two to three sessions if the work feels lopsided or destabilizing in the wrong way.

Couples often enter therapy hoping for problem resolution: we'll figure out the money fight, the sex fight, the in-law fight. The Gottman research is unambiguous that two-thirds of conflict in stable couples is perpetual. Working therapy does not eliminate these conflicts; it changes the bond's relationship to them. Couples learn to have the same fight without contempt, to recognize it as theirs, to laugh at it after, to repair quickly when it spikes. The target is not a fight-free marriage; it is a marriage that can fight without bleeding out. Couples who measure progress by problem disappearance often quit therapy that is actually working, because the fights are still there.

Most evidence-based couples models assign work between sessions: tracking interactions, practicing specific scripts, doing rituals of connection. Couples who do the homework progress faster; couples who treat therapy as a weekly venue for the same fight in front of a third party progress slowly or not at all. The mechanism of change is not the hour with the therapist; it is the six days between sessions, where the new patterns either take root or do not. Therapists who do not assign or follow up on homework are doing weaker work. Couples who do not do homework are paying for an audience.

Many couples need concurrent individual therapy to make couples work succeed. Trauma history, attachment injury, addiction, untreated depression — these load onto the couples frame and overwhelm it. A competent couples therapist will sometimes pause couples work to recommend one or both partners do individual work first, then return. Couples who resist this — "we're here for us, not for me" — often stall. The bond cannot regulate around individual material that is itself unregulated. Sequential or parallel work is not a failure of the couples model; it is the model recognizing its limits.

Some couples should not stay together, and good couples therapy will sometimes surface this. The therapist's job is not to save the marriage at all costs; it is to help the partners see clearly. When the gap is too large — values, life direction, harm history, fundamental incompatibility that has hardened — the kindest service is to help the couple separate with less damage than they would otherwise. Therapists who refuse to allow this conversation are doing harm. Couples who refuse to allow it are sometimes purchasing a slower, crueler ending. The 1,000-Page Manual's Law 0 applies hardest here: the humility to admit that not every bond is meant to last, and that meaning is not the same as failure.

A competent first session typically includes: separate intakes or at least separate moments with each partner; explicit screening for IPV, affairs, and substance use; some assessment of attachment style or cycle; and a tentative formulation of "here is what I see happening between you." A first session that is mostly the therapist explaining their philosophy, or that lets the couple dump into the room without structure, is a weaker start. Couples can usually tell within two or three sessions whether the therapist is competent. The signs are concrete: the cycle is being named accurately, the therapist interrupts destructive escalation, both partners feel heard rather than judged. Absent these, find someone else.

If you are considering couples therapy, do this: ask the therapist directly which model they primarily practice, how many couples specifically they see per week, what their training is in couples work, and what their typical course of treatment looks like. If they cannot answer these crisply, look elsewhere. Expect to pay out of pocket for the best work, because insurance billing distorts the frame. Plan for at least three to six months of weekly sessions before judging outcome. Do the homework. Accept that the bond may not survive, and that an honest answer is better than a long lie. Treat the therapist as a coach for a skill you and your partner will practice for the rest of your lives, not as a mechanic who will hand back a repaired marriage.

1. Johnson, Susan M. The Practice of Emotionally Focused Couple Therapy: Creating Connection. 3rd ed. New York: Routledge, 2019. 2. Johnson, Sue. Hold Me Tight: Seven Conversations for a Lifetime of Love. New York: Little, Brown, 2008. 3. Christensen, Andrew, Brian D. Doss, and Neil S. Jacobson. Reconcilable Differences: Rebuild Your Relationship by Rediscovering the Partner You Love—Without Losing Yourself. 2nd ed. New York: Guilford Press, 2014. 4. Christensen, Andrew, David C. Atkins, Sara Berns, Jennifer Wheeler, Donald H. Baucom, and Lorelei E. Simpson. "Traditional Versus Integrative Behavioral Couple Therapy for Significantly and Chronically Distressed Married Couples." Journal of Consulting and Clinical Psychology 72, no. 2 (2004): 176–191. 5. Gottman, John M., and Nan Silver. The Seven Principles for Making Marriage Work. New York: Crown, 1999. 6. Gottman, John M., and Julie Schwartz Gottman. Eight Dates: Essential Conversations for a Lifetime of Love. New York: Workman, 2018. 7. Doherty, William J. Take Back Your Marriage: Sticking Together in a World That Pulls Us Apart. 2nd ed. New York: Guilford Press, 2013. 8. Doherty, William J., and Steven M. Harris. Helping Couples on the Brink of Divorce: Discernment Counseling for Troubled Marriages. Washington, DC: American Psychological Association, 2017. 9. Real, Terry. The New Rules of Marriage: What You Need to Know to Make Love Work. New York: Ballantine, 2007. 10. Wile, Daniel B. After the Honeymoon: How Conflict Can Improve Your Relationship. Oakland: Collaborative Couple Therapy Books, 2008. 11. Tatkin, Stan. Wired for Love: How Understanding Your Partner's Brain and Attachment Style Can Help You Defuse Conflict and Build a Secure Relationship. Oakland: New Harbinger, 2011. 12. Perel, Esther. The State of Affairs: Rethinking Infidelity. New York: HarperCollins, 2017.