Does Couples Counselling Work

Almost every page answering this question opens with a success rate somewhere between 70 and 90 per cent. We went looking for where those numbers came from, and what we found is the reason this page reads differently from the rest.

We are Richard and Helena Boyd and we have run a couples practice in Perth since 2005. What follows is the actual evidence on whether relationship counselling works, including the parts that do not flatter our own profession. If you are trying to make informed decisions about spending money and emotional energy on this, you deserve the real numbers rather than the marketing ones.

The Short Answer

Yes, on average, and considerably less reliably than the industry suggests.

Couples therapy has genuine research support as a treatment for relationship distress. The average couple who completes it ends up better off than roughly 80 per cent of similar couples who received nothing. That is a real effect, comparable to many accepted medical treatments.

It is also true that around half of couples do not reach recovery, that a meaningful minority get worse, and that of those who do improve, only about half are still improved five years later. Both of those things are true at once, and a page that tells you only the first half is selling something.

What The Numbers Actually Say

Here is the picture from the best available synthesis of the research, stated plainly.

At the end of treatment, roughly 48 per cent of couples have recovered, about 18 per cent have improved without fully recovering, around 21 per cent are unchanged, and about 13 per cent are worse than when they started. The average effect size across controlled trials is around 0.84, which is what puts the treated couple ahead of about 80 per cent of untreated ones.

Those figures come from research trials, where therapy is manualised, therapists are supervised and couples are screened. Everyday practice looks different. In a study of 554 couples across 30 counselling centres, effect sizes ran at roughly half the trial figures, with relationship distress at 0.38 and relational functioning at 0.58. That is still a real benefit. It is not the benefit the trials advertise.

One more number worth holding. In that same practice based study, 29 per cent of couples had separated or were heading that way by follow up. Research shows a genuine effect and a genuinely wide spread of outcomes.

Where The 70 To 90 Per Cent Figure Came From

This is the part we think matters most, because the figure appears on nearly every competing page and it does not survive contact with its source.

The 70 to 75 per cent version traces to a 1999 meta analysis of Emotionally Focused Therapy covering four trials, three of which were run by the founders of that model. Four small studies of one branded approach, in 1999, generalised into a claim about couples counselling as a whole.

The 90 per cent version has a different and equally weak origin. It comes from a set of figures published by an American professional association, most likely descending from a 1996 survey in which clients were recruited by their own treating therapists. No control group, no standardised outcome measure, no follow up, and satisfaction rather than outcome. It tells you people liked their therapist. It tells you nothing about whether therapy worked.

The 80 per cent that circulates is probably a mangling of something real. The average treated couple does better than about 80 per cent of untreated couples. That is a comparison against people who got nothing, not a proportion of couples who succeed. Those are entirely different statements and only one of them is supported.

We are not making a pedantic point. If a page opens with an invented number and then asks you to trust its advice about your marriage, the number has already told you something useful about the page.

Does The Improvement Last

Durability is the question almost nobody publishes, and it is the one that should shape your expectations.

At two years, behavioural couple therapy shows about 37 per cent recovered and 25 per cent deteriorated or separated. Emotionally focused therapy shows about 39 per cent recovered and 32 per cent deteriorated or separated. At five years, roughly a third remain recovered across approaches.

Put another way, of the couples who reliably improve by the end of therapy, only about half are still in that position five years on. In applied settings, somewhere between 30 and 60 per cent relapse to where they started at some point after finishing.

That is not an argument against going. It is an argument for treating what you learn as a practice rather than a repair, and for coming back briefly when things drift rather than waiting for the next crisis.

When Couples Counselling Makes Things Worse

Being honest about this is unusual on a therapy website and it is the section we would most want you to read.

About 13 per cent of couples are worse off at the end of treatment in the largest trial. At two years that rises to between a quarter and a third who have deteriorated or separated. In routine practice, roughly half show no reliable benefit at all.

What is known about why is correlational rather than settled, but four factors come up repeatedly. Where the two people want different things from the process. Where the alliance is split, meaning one partner feels the counsellor is on their side and the other does not. Where pretreatment distress is extremely high. And where an affair remains undisclosed, which is the one subgroup in the research that reliably got worse rather than better.

There is also a factor nobody measures well, which is that couples often arrive late. The average is 2.68 years from the problems starting, and by then some patterns have set hard. That is not a reason to stay away if you have waited, it is a reason not to wait longer if you have not.

One caution on interpreting all this. Separation is counted as a poor outcome in the data, and it is not always a poor outcome in life. Some couples separate through therapy and both people are better for it, which the numbers record as failure.

The Australian Context

Worth grounding this in local figures rather than the American ones most pages import.

In 2025 there were 49,158 divorces granted in Australia, a crude divorce rate of 2.2 per 1,000 people. The median time from marriage to separation was 9.4 years, and from marriage to divorce 13.4 years. The divorce rate has been broadly flat across the decade, though divorce is happening later in life than it used to.

You will see it claimed that 40 to 50 per cent of first marriages end in divorce. That is an American figure, contested even there, and it is not derivable from Australian data. The most recent Australian lifetime estimate, published in 2007, projected that around a third of marriages entered into in the early 2000s would end in divorce. It was a projection rather than a count, and it is now old.

Avoid the other common trick too, which is dividing this year’s divorces by this year’s marriages to get one in three. Those are two unrelated groups of people and the sum means nothing.

The Methods, And What Each Is Actually Worth

Various methods get named on therapy websites as though the label guarantees the outcome. Here is what the evidence behind each actually looks like.

Emotionally Focused Therapy

Emotionally focused therapy is an attachment based approach that works on the negative cycle between partners and rebuilds emotional connection by helping each person express the vulnerable material underneath their usual position.

It has the strongest research base of the branded models, and it is also the one most inflated in marketing. The founding meta analysis of four trials produced an effect size around 1.31, with three of those four trials run by the model’s developers. An independent German review of twelve studies produced 0.73. Both are real, and the gap between them is what allegiance effects look like in practice.

The Gottman Method

This is where the biggest overclaim in the field sits, and it needs separating carefully.

John Gottman’s observational research on how couples actually interact is a large, genuinely influential body of work. The Gottman Method as a treatment protocol is a different thing. We could not identify a single published, peer reviewed, independent randomised trial of it as a therapy for relationship distress. What exists is a dismantling study of a weekend workshop, a self described pilot study, an unpublished doctoral dissertation, and one non Gottman study using a matching design to test a psychoeducation programme rather than the therapy.

That does not make it useless. The concepts are clinically sensible and we use some of them. It means that describing the method as proven borrows credibility from the research without earning it.

Behavioural Couple Therapy

The original behavioural approach, which increases positive exchanges between partners and directly trains communication and problem solving skills.

It has the largest and most replicated evidence base of anything in the field, with effect sizes around 0.84 across approaches and independent replication across the United States, Europe and Australia. Less fashionable than the attachment models, and better tested.

Integrative Behavioural Couple Therapy

An extension of the behavioural approach that adds acceptance work, meaning helping two people live with differences that are not going to change rather than only trying to change them.

It is anchored by the largest single trial in the field, 134 couples with two year and five year follow up. Head to head against traditional behavioural couple therapy it showed no significant difference at the end of treatment, which is itself an informative result.

Cognitive Behavioural Couple Therapy

Behavioural work plus explicit attention to what each partner thinks, meaning the assumptions, standards and interpretations that shape how a given event lands.

The evidence supports it for relationship distress. Notably, adding the cognitive components has not demonstrated benefit over the behavioural components alone, which is a recurring pattern in this literature.

Why No Method Comes Out On Top

If you take one thing from the section above, make it this, because it changes how you should choose.

Head to head comparisons of couples therapy approaches rarely show any of them superior. The therapeutic techniques differ, the underlying theories differ, and the outcomes come out roughly level. What the approaches share, and what most likely does the work, is a structured process for slowing down the same arguments, examining relationship patterns rather than incidents, and making it possible to say difficult things in a room where both people stay.

The practical implication is that shopping for a brand is a poor use of your attention. What the common elements suggest is that the person and the process matter more than the acronym on their website.

What Actually Changes When It Works

Success rates are abstract. Here is what the change looks like from inside the room, since that is what you are actually buying.

The first thing that shifts is usually the speed of an argument. A couples counselling session takes one specific incident from the week and slows it down until each of you can see where it turned, which is normally much earlier than either of you thought. Doing that repeatedly is how relationship dynamics become visible to the people inside them. You start catching the pattern while it is running rather than reconstructing it afterwards.

The second thing is what can be said. Most couples arrive with a list of things that cannot be raised at home without a fight, and much of the therapy process is simply making a safe space where those things get said once, properly, with someone present who keeps the conversation from collapsing. Improved communication in practice tends to mean fewer conversations that end badly rather than more conversations overall.

The third is emotional connection, and it is usually last. Where communication issues have been running for years, warmth returns after the arguments settle rather than before, and intimacy generally follows that rather than leading it. Couples who expect the affection back first often conclude too early that nothing is working.

None of that requires a personality transplant. What a healthy relationship looks like after therapy is usually the same two people with the same differences, handling them at a fraction of the cost.

Does It Help Mental Health As Well

This is where couples counselling has evidence most people are unaware of, and also where the overclaiming starts again.

For depression, the evidence is strong enough to be in clinical guidelines. The UK’s National Institute for Health and Care Excellence recommends behavioural couples therapy for adults with depression where relationship problems may be contributing, or where involving the partner may help, delivered over 15 to 20 sessions across five to six months. A Cochrane review of 14 trials covering 651 participants found couple therapy about as effective as individual therapy for depressive symptoms, with an additional benefit for relationship distress, while rating the certainty of that evidence as low.

For anxiety, the honest answer is that the evidence is not there yet. There is no meta analysis of couple therapy for anxiety disorders as a class. Couple based work for anxiety, obsessive compulsive disorder and post traumatic stress shows promise, but direct comparisons with individual therapy remain limited. Anyone telling you couples counselling significantly reduces anxiety is ahead of the research.

There is a broader point underneath. Relationship distress has a negative impact on individual mental health, and improving the relationship often improves how each person is doing. In the largest naturalistic study, psychological distress improved more than relationship satisfaction did.

Couples Therapy And Substance Use

Most pages say substance use must be treated before couples therapy can work. That is wrong, and the correction is useful.

Behavioural Couples Therapy for alcohol and drug use disorders has a solid evidence base as a treatment for the substance use itself. Across 12 studies covering 754 couples it outperformed individual based treatment with an effect size of 0.54, and a more recent analysis of 19 trials found it superior on substance use, consequences of use and relationship satisfaction at both post treatment and 12 months. It is recommended in UK clinical guidelines for harmful drinking and mild alcohol dependence, and for stimulant or opioid misuse.

Two qualifications matter. This is a manualised, substance focused protocol delivered with a non using partner, not general relationship counselling. And the trials do not cover acute intoxication, unmanaged withdrawal, or both partners using, so those situations still need individual therapy or specialist treatment first.

What has no evidence at all is generic couples counselling with an actively drinking or using partner, which is what the usual advice was clumsily pointing at.

Does It Work After An Affair

The data here is genuinely two sided and worth knowing before you decide.

Couples who come to therapy after infidelity arrive more distressed than other couples and improve faster, finishing treatment at a comparable level. That is the encouraging half, and it holds across both a controlled trial and a large community sample.

The five year picture is harder. Couples who presented with an affair divorced at 43 per cent against 23 per cent for other couples, and that gap remained even after excluding couples where the affair had stayed secret.

One factor stands out. Couples where the affair remained concealed deteriorated rather than improved. That is a finding about disclosure of the affair itself, not about indiscriminate honesty regarding every thought and detail, which can deepen the injury rather than repair it. Structured approaches work through the immediate impact first, then how the affair came about, then what happens next.

What Predicts Whether It Works For You

The research on this is more useful than the general success rates, because it is about your situation rather than the average person.

The clearest signal is whether both of you want the same thing. In a study of 249 couples, 7.8 per cent had separated six months later where both partners came wanting to improve the relationship. Where their goals differed, 45 per cent had separated. Where both came to work out whether to continue at all, 56 per cent had.

Beyond that, what a couple looks like at the initial session predicts outcome poorly. Higher initial distress, low mutual support and elevated depression are associated with worse results, but none of it predicts reliably at the individual level. What does predict well is whether anything starts shifting early. In one study, change by the end of therapy correctly classified couples as stable or separated 89 per cent of the time.

That fits what the timing research shows. Measurable improvement tends to happen across roughly the first four sessions and then plateau, and about a third of couples are already off track by the third session. This is why we ask directly whether it is helping rather than assuming that more sessions produce more change.

What Makes A Good Couples Therapist

Two common pieces of advice here are weaker than they sound, and one thing matters more than either.

You will read that you should choose someone with several years of experience. General years in practice is one of the weakest predictors in the entire psychotherapy literature. In a study of 158 couples across 18 therapists, therapists accounted for about 8 per cent of the variance in outcomes. Within that 8 per cent, experience specifically with couples explained about a quarter, and the therapist’s gender and professional discipline predicted nothing at all.

What explained about half of that therapist effect was the quality of alliance they typically formed with clients. And there is a couple specific finding that matters more than the general one. Across 48 studies covering more than 1,500 couples, the alliance was associated with outcome at around 0.30, and split alliances predicted poorer outcomes at a similar strength. Where one partner feels the counsellor is theirs and the other does not, the work goes badly.

So the useful question is not how senior someone is. It is whether they work with couples routinely, whether both of you feel they are genuinely your therapist, and whether they track whether the work is helping. If it is not right after a few sessions, a different therapist is a reasonable move rather than a failure.

Choosing Between Your Options In Perth

There are several routes into this locally, and the differences are worth understanding before you book.

Relationship counsellors and couples counsellors in private practice are the most common option. Counselling and psychotherapy are not registered professions in Australia, so what you can check is registration on the ARCAP register and membership of PACFA or the Australian Counselling Association, along with the level. Some people search for a relationship psychologist specifically, which means an AHPRA registered psychologist. Worth knowing that professional discipline did not predict outcomes in the research, so this is less decisive than it sounds, and there is no Medicare rebate for couples counselling regardless of who you see.

Relationships Australia WA offers counselling on a sliding scale based on household income, at locations including Fremantle, Joondalup, Midland, Mandurah and Gosnells. Where cost is the barrier, that is the first place to look rather than not going.

Whichever route you take, a few questions help. Ask what approach they use and why. Ask whether they work with couples routinely or occasionally. Ask how they will know whether it is helping. And notice, by the second session, whether both of you feel they are your therapist rather than one of yours. Most practitioners will talk you through all of that on the phone before you book, and a reluctance to do so is itself informative.

Most practitioners will take a brief history of the relationship and some family history from each of you early on, since how conflict was handled in each of your families tends to explain more than either of you expects. If a counsellor shows no interest in either, or promises to fix specific issues in a set number of sessions, that is worth noticing.

Where Couples Counselling Is Not The Answer

Some situations call for something else, and one of them is a safety matter.

Couples counselling is generally not appropriate where there is family violence or coercive control. Australian professional guidance is that joint counselling is usually contraindicated, because you cannot speak freely in front of someone using control against you, and what you say in the room can be used against you afterwards. Individual support is the safer starting point. If this is your situation, contact 1800RESPECT on 1800 737 732, the Women’s Domestic Violence Helpline on 1800 007 339 or the Men’s Domestic Violence Helpline on 1800 000 599, or 000 in an emergency.

Beyond that, individual therapy usually needs to come first or run alongside where one person’s untreated depression, addiction or childhood trauma is driving most of what happens between you. Individual sessions can also be the right setting where sexual difficulties have a physical cause that needs a doctor, or where one partner is not willing to attend at all.

External pressure is a different case. Financial strain, a life transition, extended family conflict or the arrival of children put pressure on relationships that were previously fine. There is reasonable evidence that couples can be helped to cope with outside stress together rather than separately, though the programmes tested are preventive rather than treatments for a crisis, and the effects are modest.

Relationship Diversity

A brief note, because most pages on this subject assume one relationship structure.

Around one in five single Americans report having been in a consensually non monogamous relationship at some point, and no equivalent Australian figure has been published. A 2025 meta analysis of 35 studies covering 24,489 people, led from Australian Catholic University, found no meaningful difference in relationship satisfaction or sexual satisfaction between consensually non monogamous and monogamous people.

We will be straight about the limits of what is known. There are no outcome trials of relationship therapy with polyamorous relationships or other non monogamous structures, so nobody can quote you effectiveness data. What the research does document is stigma, including from therapists. The assumption that these relationships are inherently less satisfying is not supported by the evidence.

The Australian Evidence Gap

Worth saying plainly, because it affects how much weight to give everything above.

There is no published Australian randomised trial of couples counselling. The only substantial Australian evaluation is now more than thirty years old and reported descriptive percentages without a comparison group. Relationships Australia publishes service wide outcome data, but it covers all family and relationship services rather than couples specifically and has no control group.

The effectiveness evidence we have relied on throughout this page comes from the United Kingdom, Germany, the United States and Switzerland. There is no obvious reason it would not transfer, and it has not been tested here.

What This Looks Like In Practice

These are composite examples drawn from our couples work, with identifying details changed.

A couple in their forties came in with the same arguments running for years and no idea what was underneath them. Six sessions in, one of them said something they had never said in eighteen years. That made a huge difference, and it was not a technique that produced it so much as a room where it could be said.

A couple in their thirties arrived with genuinely different goals, one hoping to repair and one already gone. Naming that in the first session, rather than six sessions later, changed what the work was for and probably saved them both a great deal of pain.

A man in his fifties came alone after his wife declined to attend. Working on his own side of a pattern shifted the pattern anyway, which is a more common outcome than people expect.

Frequently Asked Questions

The questions people ask us most often about this.

What Is The Real Success Rate For Couples Therapy?

Around half of couples recover by the end of treatment, roughly a fifth improve without fully recovering, a fifth are unchanged and about one in eight are worse. At five years, roughly a third remain recovered. Any page quoting 70 to 90 per cent is repeating a figure that does not hold up.

How Do You Know If It Is Working?

Early, and by asking. Most measurable change happens across the first four sessions and then plateaus, and couples off track by session three benefit considerably less. If nothing has shifted in the conversations by then, that is worth raising directly with your counsellor rather than waiting.

Is It Worth Trying With A Different Therapist If The First Did Not Help?

Often yes. Therapists account for about 8 per cent of outcome variance, and split alliances predict poor results. If one of you never felt the counsellor was theirs, that is a specific and fixable problem rather than evidence that counselling does not work for you.

Can Couples Counselling Help If We Are Probably Separating?

Yes, and that is a legitimate use of it. Deciding clearly, and separating with less damage where children or shared life are involved, is a real outcome. The research counts separation as failure, which is a limitation of the research rather than a description of your situation.

How Many Sessions Will We Need?

There is no standard. In the largest study of routine practice the average was 23 sessions, with a range from two to 150. What matters more than the total is that something starts moving early, so treat the first handful as the period in which you both decide whether this is the right fit for the future.

Does It Work If Only One Partner Attends?

It can. Individual counselling about relationship issues is common and often enormously helpful, because patterns need two participants and they change when one person stops running their half. It is a different perspective on the same problem rather than a lesser version of it.

Getting Started In Perth

So, does couples counselling work. On the evidence, yes for most couples, unreliably, and considerably less spectacularly than the industry claims. About half recover, a meaningful number do not, gains need maintaining, and the method matters less than whether both of you can be honest with the person in the room.

We would rather you knew that before booking than discovered it afterwards. A practice that oversells its results is a practice you cannot calibrate, and calibration is what lets you tell early whether something is helping. Realistic expectations are not pessimism, they are the thing that stops a couple abandoning useful work at session four because the transformation they were promised has not arrived on schedule.

What we can say from twenty years of this work is that the couples who do well are rarely the ones with the smallest problems. They are the ones who arrive wanting the same thing, who say the difficult thing sooner rather than later, and who treat what they learn as an ongoing practice in daily life rather than a course they completed. Open discussions get easier with repetition, and better communication skills only hold if they survive a bad week.

If the same conversation keeps happening at home, or has quietly stopped happening, that is usually the point at which this is most workable and the point at which people most often wait. Coming early is easier work than coming late.

We see couples at our Inglewood rooms and online across Western Australia, evenings and Saturday mornings included, with no referral needed. Call us on 1300 956 227, book a session, or read more about what to expect from couples counselling and our couples counselling services.

This page is general information and is not professional medical advice or a diagnosis. If you are distressed, contact Lifeline on 13 11 14, or 000 in an emergency.