What should a separation agreement cover?

Length, where each person lives, how money is handled, where the children sleep and how they move between households, how often the two of you are in contact, whether either of you sees other people, and what happens when the period ends. The negotiation is as informative as the document.

Does separating make divorce more likely?

Most separations do end in divorce, but the direction of that relationship is not established. Couples who separate are already in worse shape than couples who do not, so the outcome partly reflects who separates rather than what separating does.

Can you date other people during a trial separation?

That is exactly the kind of question the written terms exist to settle in advance. Clinical practice has generally treated seeing other people as incompatible with a separation intended to test whether the marriage can be repaired, but the point is that both people agree the rule beforehand rather than discovering it in an argument.

Is there an alternative if we do not want to live apart?

Discernment counselling was designed for couples where one partner is leaning towards ending it and the other is not. It is short, aimed at reaching a decision rather than at repair, and requires nobody to move out.

DivorceLimited evidence

Should you try separation before divorce?

The short answer

Sometimes, but not for the reason it is usually recommended. Around a third of married couples who separate reconcile at least once, and about a third of those reconciliations break down again inside a year. What almost no study measures is the specific thing you are asking about: a separation with written terms, a deadline and counselling attached. The clinical literature on that version is small, mostly from the 1970s and 80s, and has never been tested against a comparison group.

14 min readPublished July 28, 2026Updated July 28, 2026

What readers think

When someone proposes a trial separation, what is it usually?

Anonymous. One vote per reader.

Two different things travel under the same name, and telling them apart decides the answer. One is a couple who agree to live apart for a fixed period, with terms set in advance about money, children, contact, and whether either of them sees other people, while both keep attending counselling. The other is one person moving out because the situation has become unbearable, with nothing agreed and no end date.

Only the first is a trial. Nearly every number you will be quoted comes from the second, because a national survey can see who stopped sharing an address and cannot see whether there was a plan.

The numbers describe separations, not trials

Among married couples who separated, roughly a third reconciled at least once, and about a third of those reconciliations broke down again within a year[1]. Survey work from the same period found that around 44 percent of separated women attempted a reconciliation at all, and that about one attempt in three held[2].

More recent data on people who separate after age 50 is considerably bleaker. Within ten years, about 7 percent of women and 11 percent of men had reconciled, while 81 percent of women and 63 percent of men were still separated and had not repartnered[3]. Later-life separation looks less like a pause and more like a slow, undeclared ending.

Read carefully, none of these figures are about trial separation. They count everyone who stopped living together, which includes the couple who split during a crisis and the couple who quietly ran out of road. So the defensible summary is that most separations do not end in reconciliation, and that this tells you remarkably little about a separation you would design deliberately.

Structured separation is a clinical idea with very little outcome evidence

The version therapists sometimes recommend has a name and a shape. A structured or controlled separation sets a length, usually a few months, and a written agreement covering where each person lives, how money works, how the children move between households, how much contact the couple has, whether either of them dates, and a commitment that both stay in counselling for the duration.

The idea is not new. It was described in the clinical literature in 1972, with a one-year follow-up on eighteen couples who completed a three-month structured separation with counselling[4], and set out again as a treatment and decision-making framework a decade later[5]. Practitioners have been refining the model ever since.

What has not happened is testing. There is no randomised trial, no comparison group of couples who separated without terms, and no large sample. The case series are decades old and number in the tens. That does not make the approach wrong. It means anyone quoting a success rate for controlled separation is quoting something that was not measured.

The design is where the risk sits

The failure mode of an unstructured separation is drift. Without an end date, the arrangement stops being a decision-making period and becomes the arrangement. Each of you builds a life that runs without the other, and the question the separation was meant to answer gets postponed rather than examined.

The fragility of reconciliation points the same way. A meaningful share of couples who do get back together separate again quickly[1], which suggests that returning without having changed anything reproduces the original problem on a delay.

There is also an asymmetry worth naming. A separation proposed by one partner as a way to think can be experienced by the other as a departure that has already happened, and it is sometimes used that way, as a softer exit that avoids saying the word. If the two of you cannot state out loud what the separation is for, the terms will not fix that, but the attempt to write them will expose it.

Separation also changes facts on the ground. Housing, finances, tax, and the pattern of who the children live with all start setting a precedent, and in many jurisdictions the arrangements you fall into informally influence what a court later treats as normal.

Discernment counselling is the better evidenced version of the same instinct

The impulse behind a trial separation is usually the need for room to decide rather than the need to live apart. There is a clinical approach built for exactly that situation, and it has more behind it than the separation literature does.

A survey of 2,484 divorcing parents who had already taken the required parenting class found that about one in four individuals still believed the marriage could be saved with hard work, and in around one couple in nine both partners did[6]. Interest was far higher among people whose spouse had initiated the divorce, which is a reminder that these decisions are rarely being made jointly.

Discernment counselling was developed for couples where one partner is leaning out and the other is leaning in, on the observation that ordinary couples therapy assumes a shared goal that these couples do not have. In an analysis of 100 consecutive cases, roughly half chose to enter couples therapy with the aim of reconciling, and most of the rest chose the divorce path[7]. It is short, it is explicitly about deciding rather than repairing, and it does not require anybody to move out.

It depends, on what?

What changes the answer

The uncomfortable part is that the evidence here cannot answer the question in the form you are asking it.

The large surveys measure separation as an event, and they are dominated by separations that were the beginning of a divorce rather than an attempt to avoid one. The clinical work describes the thing you actually mean but rests on tiny, old, uncontrolled samples. Nobody has run the study that compares couples who separated with agreed terms against couples in the same state who did not, and until somebody does, claims in either direction are inference rather than evidence.

Selection makes it worse. Couples who can negotiate a written separation agreement, keep to it, and both stay in counselling are already demonstrating capacities that predict a better outcome regardless of where they sleep. If structured separations do show better results, some of that is the couples, not the structure.

There is one thing the pattern does support, weakly but consistently. Separations that were never designed mostly do not end in reunion, and reunions that were not preceded by change frequently do not hold. That is an argument for being specific about what the separation is for, what would have to be different for it to end in reunion, and when you will look at it again. It is not an argument that separating first improves your odds, because that comparison has not been made.

Anyone offering you a percentage for whether trial separations work is not describing a finding. That number does not exist.

Where people disagree

Both sides of it

Reasonable people land in different places on this. Here is the strongest version of each case, not a straw man of the one we disagree with.

The case for yes

A fixed term with written conditions forces the couple to state what the separation is for, which is a conversation many have been avoiding for years.
Time apart interrupts a daily pattern that both people may be too depleted to change while living inside it, and some couples genuinely cannot think clearly at close range.
It preserves options that divorce closes. Reconciliation after a separation is uncommon but possible, whereas reconciliation after a completed divorce is rarer still.
For couples where one partner has already decided and the other has not, a defined period is more honest than an indefinite stalemate, and it puts a date on a decision that has been drifting.

The case for no

The evidence that structured separation improves the odds does not exist. The case series are tiny, decades old, and have no comparison group.
Most separations end in divorce rather than reunion, and a separation is far more often the first step out than a pause on the way back.
Without a deadline, the arrangement becomes permanent by default and the decision it was meant to force never gets made.
Living apart is expensive, and for many couples running two households is the thing that makes divorce financially unavoidable rather than optional.

Evidence

What the research says

Moderate evidenceA single long-running panel of young adults with complete union histories
Among married couples who separated, roughly a third reconciled at least once, and about a third of those who reconciled separated again within a year.
What this does not show
Drawn from a panel of White young adults followed from age 15 to 31, so it does not describe older couples, longer marriages, or other populations. Separations are counted as events without any information about whether terms, a deadline, or counselling were involved.
Binstock, G., & Thornton, A., Separations, Reconciliations, and Living Apart in Cohabiting and Marital Unions (2003)(opens in a new tab)
Moderate evidenceNational Survey of Families and Households
Around 44 percent of separated women attempted a reconciliation, and roughly one attempt in three resulted in a lasting reunion. Attempts were more common among women with fewer economic resources.
What this does not show
Based on United States survey data collected in 1987 and 1988, so the marital and economic context is now dated. Self-reported and retrospective, and reconciliation is measured only from the wife's account.
Wineberg, H., Marital Reconciliation in the United States: Which Couples Are Successful? (1994)(opens in a new tab)
Moderate evidence1,075 adults aged 50 and over followed in the Health and Retirement Study
Among people who separated after age 50, about 7 percent of women and 11 percent of men had reconciled within ten years, while 81 percent of women and 63 percent of men remained separated without a new partner.
What this does not show
Restricted to adults over 50, where separation dynamics differ sharply from those of younger couples. The authors note that reconciliations occurring and ending between survey waves are invisible to the design, which likely understates the true rate.
Brown, S. L., Lin, I.-F., Marino, F. A., & Mellencamp, K. A., Marital Separation, Reconciliation, and Repartnering in Later Life (2025)(opens in a new tab)
Limited evidence18 couples followed for one year
Structured separation with counselling, a three-month agreement with defined terms and continued therapy, was described clinically with a one-year follow-up on the couples who completed it.
What this does not show
Eighteen couples, no comparison group, no randomisation, and published in 1972. It documents that the approach was applied and followed up, not that it outperforms separating without terms or not separating at all.
Toomim, M. K., Structured Separation with Counseling: A Therapeutic Approach for Couples in Conflict (1972)(opens in a new tab)
Moderate evidence2,484 divorcing parents
About one in four individual divorcing parents still believed the marriage could be saved with hard work, and in roughly one couple in nine both partners did. Interest was much higher among those whose spouse had initiated the divorce.
What this does not show
Surveyed parents in one United States jurisdiction after a mandatory parenting class, so the sample is parents in an active divorce process rather than couples considering separation. Self-reported attitudes at a single moment, not a prediction of what those couples did next.
Doherty, W. J., Willoughby, B. J., & Peterson, B., Interest in Marital Reconciliation Among Divorcing Parents (2011)(opens in a new tab)

How we rate evidence strength

Strong evidence: Multiple independent studies, including replications or meta-analyses, point the same way.
Moderate evidence: Several studies support this, but samples are limited or findings vary by population.
Limited evidence: Early, small, or single-study evidence. Treat this as a reasonable hypothesis, not a settled fact.
Contested: Credible researchers disagree, or the evidence points in conflicting directions.

Comparative view

Different perspectives

Traditions and disciplines answer this differently. We describe what each one teaches. We are not telling you which is right, and no tradition speaks with a single voice.

What the Evidence Supports

The strongest claim available is descriptive: most marital separations do not end in reconciliation, and this holds across several decades and datasets. The claim that a designed separation improves the odds is not supported, and it is also not refuted, because the comparison has not been run.

The reconciliation figures also drift substantially with age and cohort, from around a third in a young adult panel to roughly one in ten after age 50, which is a warning against quoting any single number as the rate. The discernment counselling evidence is stronger in that it reports outcomes on a defined clinical population, but it too lacks a control group.

Therapist Perspective

Clinicians who use structured separation are generally careful to distinguish it from moving out. The written agreement is treated as the active ingredient, because the process of negotiating it reveals whether two people can still cooperate under pressure, and because the terms prevent the arrangement from quietly becoming permanent.

A common practitioner observation is that couples asking for a trial separation are often asking for permission to stop trying without saying so, and that the request itself is worth examining before the logistics are. Therapists also tend to warn that a separation with no counselling attached is not the intervention, it is just an absence.

Secular Perspective

A secular framing treats separation as a practical instrument rather than a moral stage, to be judged on whether it serves the people involved. On that view there is no independent virtue in having tried separation first, and no obligation to exhaust every intermediate step before ending a marriage that both people know is finished.

What the framing does emphasise is honesty about intent. Proposing a separation you already intend to make permanent, without saying so, imposes a cost on someone who is spending months planning around a decision that has already been taken.

Islamic Perspective

Mainstream Sunni and Shia jurisprudence

Islamic law contains staged mechanisms that resemble a structured separation more closely than the informal Western version does. Arbitration involving a representative from each family is prescribed in the Qur'an when a breach between spouses is feared, and the waiting period following a revocable divorce keeps the marriage legally intact while the couple decides, with rules on residence and maintenance that are set rather than negotiated.

Jurists have also consistently treated leaving a wife suspended, neither properly married nor released, as a wrong rather than a neutral pause, which frames an open-ended separation as a harm to be remedied rather than a reasonable holding position.

Not a settled view

Schools differ on the length and conditions of the waiting period, on the grounds and waiting times after which a wife may seek judicial dissolution for desertion or absence, and on how far family arbitration should be binding. Contemporary scholars also disagree about how classical procedures should map onto countries with codified civil family law.

Practical steps

What you can actually do

Write the terms down before anyone moves. Length, money, where the children sleep and how they move, how often the two of you talk, whether either of you sees other people, and what happens at the end. An unwritten separation is not a trial.
Set a date and put a review in the middle of it. Three months is the length the clinical model has always used, and open-ended is the version that fails.
State what would have to be different for the separation to end in reunion. If neither of you can name anything specific, you are not testing a hypothesis, you are waiting.
Keep counselling running for the whole period. In the model this idea comes from, the counselling is not an optional extra, it is the mechanism.
Ask for discernment counselling if one of you is leaning out. It is short, it is designed for exactly that mismatch, and it does not require anyone to leave the house.
Take legal advice on what moving out does to your position before you do it, and do not assume the informal arrangement you fall into will stay informal.

Worth unlearning

Common misconceptions

That there is a known success rate for trial separation. There is not. The surveys measure separations of all kinds, and the clinical work on designed separations has no comparison group.
That reconciliation after separation is common. Around a third of separated married couples reconcile at least once, and after age 50 the figure is closer to one in ten.
That getting back together means it worked. About a third of reconciliations break down again within a year, and returning without changing anything reproduces the original situation.
That separation is a neutral pause. It shifts housing, money, routine and the children's pattern, and those arrangements tend to become the reference point for whatever follows.
That living apart is the only way to get thinking room. Discernment counselling was built for that need, has better evidence behind it, and requires nobody to move out.

The short version

Key takeaways

Roughly a third of married couples who separate reconcile at least once, and about a third of those reconciliations fail again within a year.
Research
The surveys count separations of every kind, so they say almost nothing about a separation undertaken deliberately as a trial.
Research
Structured separation with written terms and continued counselling is a clinical model from the 1970s that has never been tested against a comparison group.
Clinical consensus
Separations without a deadline tend to become permanent by default rather than resolving the question that prompted them.
Clinical consensus
Discernment counselling addresses the same need for deciding room, has more supporting data, and does not require either partner to move out.
Research

Worth saying

When to get professional help

If there is violence, intimidation, or coercive control in the relationship, this framing does not apply and a trial separation is not the right instrument. Leaving is the period of highest risk in abusive relationships, which is a reason to contact a domestic abuse service for safety planning rather than to arrange a separation privately and alone.

Take legal advice before you move out, not after. Separation alters your financial and custody position in ways that vary enormously by jurisdiction, and informal arrangements have a way of hardening into the default a court later works from. This is true whether or not you end up divorcing.

If what you want is room to decide, ask about discernment counselling by name. It was built for couples where one partner is leaning out, and standard couples therapy handles that situation badly because it assumes you both want the same outcome.

If you do separate, a couples therapist or family mediator during the separation is closer to how the approach was designed than separating and waiting to see how you feel. Individual therapy is also reasonable, particularly if you are the one who is unsure.

Find support services near you

Still wondering

What other people say

These are readers describing their own relationships, not professionals. Take them as experience, not advice, and please do not name anyone.

No one has shared their experience yet. If you have been here, you would be the first, and probably the most useful thing on this page.

Check our work

Sources

[1]
Separations, Reconciliations, and Living Apart in Cohabiting and Marital Unions(opens in a new tab)
Binstock, G., & Thornton, A. · Journal of Marriage and Family · 2003
[2]
Marital Reconciliation in the United States: Which Couples Are Successful?(opens in a new tab)
Wineberg, H. · Journal of Marriage and the Family · 1994
[3]
Marital Separation, Reconciliation, and Repartnering in Later Life(opens in a new tab)
Brown, S. L., Lin, I.-F., Marino, F. A., & Mellencamp, K. A. · Journal of Marriage and Family · 2025
[4]
Structured Separation with Counseling: A Therapeutic Approach for Couples in Conflict(opens in a new tab)
Toomim, M. K. · Family Process · 1972
[5]
Structured Separation for Marital Treatment and Decision-Making(opens in a new tab)
Granvold, D. K. · Journal of Marital and Family Therapy · 1983
[6]
Interest in Marital Reconciliation Among Divorcing Parents(opens in a new tab)
Doherty, W. J., Willoughby, B. J., & Peterson, B. · Family Court Review · 2011
[7]
Discernment Counseling for Mixed-Agenda Couples(opens in a new tab)
Doherty, W. J., Harris, S. M., & Wilde, J. L. · Journal of Marital and Family Therapy · 2016

divorce separation counselling reconciliation decision making