How Long Should a Relationship Break Be?

12 min read · Reviewed against current psychological research

You want a number. Two weeks, a month, ninety days, something you can hold onto and organise yourself around, because right now everything is uncertain and a rule would be a mercy. So you go looking, and the internet obligingly gives you numbers. Confident ones. “Two to four weeks.” “One month for every year you’ve been together.” “No more than six months.”

I’m going to do something slightly annoying and tell you the truth instead, which is that nobody knows, because nobody has studied it. And then I’m going to tell you what the research does say, which is more useful than a number anyway.

There is no evidence-based number

Let’s be direct about this. There is no body of research establishing an optimal length for a relationship break. It hasn’t been investigated. There’s no study comparing three-week breaks with three-month breaks and measuring what happened.

So when an article tells you confidently that a break should last thirty days, that number came from nowhere. Someone made it up, and then other people repeated it, and now it’s on forty websites and looks like a fact.

I’d rather tell you that than hand you a fake certainty. Especially because, as it turns out, the length isn’t the thing that decides the outcome.

What we do know, and it isn’t encouraging

What has been studied, extensively, is what researchers call relationship cycling or churning: couples who break up and get back together, sometimes repeatedly. And it’s common, nearly half of young adults report a breakup and reunion with their current or most recent partner.

The findings are consistent and they’re not cheerful. Compared with couples who don’t cycle, on-off partners report:

• Higher conflict
• More uncertainty about the relationship’s future
• Lower relationship satisfaction
• Lower commitment
• Less explicit decision-making
• Greater risk of relationship distress overall

There’s also a harder finding worth reporting honestly rather than hiding: research on churning has found it associated with higher rates of physical violence and verbal abuse in young adult relationships. That’s an association across a population, not a prediction about you, and it doesn’t mean a break causes those things. But it does mean the on-off pattern is not the harmless pause button our culture treats it as.

A break is not neutral. It’s a real intervention with a real evidence base, and the evidence is mixed at best.

The finding that actually answers your question

Here’s the useful part. Sarah Halpern-Meekin, who has done much of this research, puts the crucial insight in three words: decide, don’t slide.

Cyclical couples are specifically characterised by lower explicit decision-making. They don’t decide to get back together. They slide back, because they’re lonely, because the separation hurts more than the relationship did, because they saw a photo, because it’s easier.

And the thing that distinguishes a reunion that works from one that doesn’t isn’t how long you were apart. It’s whether the thing that caused the break has actually been addressed.

So here is the question that replaces “how long should this be?”:

What will be different?

If nothing has changed, you are not returning to a repaired relationship. You are returning to the identical relationship, plus a few months of accumulated resentment, plus fresh evidence that this is simply what the two of you do. That’s not a fresh start. That’s a lap.

What a break actually needs

Since the duration isn’t the variable, here’s what is. A break that has any chance of helping needs four things, and most breaks have none of them.

1. A stated purpose. “I need space” is not a purpose, it’s a feeling. A purpose is specific: I need to work out whether I can live with how little you want to see my family. I need to find out whether I’m staying because I want to or because I’m scared. If neither of you can say what the break is for, it isn’t a break. It’s a drift.

2. Explicit rules. Are you in contact? How much? Are you seeing other people? Is that sleeping with other people? Who are you telling? What happens if one of you wants to talk before the end date? Write it down, actually. The uncertainty is not a neutral background condition, it’s the specific thing that makes cycling relationships miserable, and you can eliminate most of it in one uncomfortable conversation.

3. An end point. Not because the length matters in itself, but because open-ended breaks don’t end, they dissolve. Nobody ever says “so that’s over then.” You just gradually stop, over months, in a fog, without either of you having to be the one who said it.

4. A decision rule. Agree in advance how you’ll actually decide. What would a yes look like? What specifically would need to have changed? What would tell you the answer is no? Deciding this while you’re calm is far easier than deciding it on the phone at midnight on day 29.

So, how long, really?

I’ll give you my reasoning, and I’ll flag clearly that this is reasoning rather than a research finding, because I’ve just spent several paragraphs criticising people who don’t.

Long enough to stop being flooded. Days are not enough. In the immediate aftermath of a bad patch you’re physiologically activated and thinking badly; a few days apart just lets you both rehearse your case. You need long enough to genuinely settle and to actually do something different, start the therapy, have the conversation with yourself, live alone and see how it feels.

Short enough to still be a break. Past a certain point, an ambiguous separation stops being a break and becomes a relationship you’re both quietly in the process of leaving. If the end date keeps moving, that’s information.

In practice, for most people, that lands somewhere in the range of several weeks to a couple of months, with a fixed date, rather than “let’s see how it goes.” But treat that as a reasonable default rather than a finding, because that is exactly what it is.

The uncomfortable possibility

One more thing, said gently, because someone reading this needs to hear it.

Sometimes a break is not a break. Sometimes it’s a breakup being administered slowly, by someone who hasn’t got the words yet, or who wants to leave without being the villain, or who is testing whether they miss you. That’s usually not cruelty. It’s often just cowardice, or genuine confusion, and most people who do it don’t fully know they’re doing it.

But if you’re the one who was told there’d be a break, you’re allowed to ask the question directly: “Is this a break, or is this how you’re ending it?” You are allowed to want an answer rather than a limbo. And if the answer keeps being postponed, that is, in itself, the answer.

The takeaway

There’s no correct number of weeks, and I’d rather say so than invent one. What the research actually points at is this: breaks fail when people slide back together to stop the pain of being apart, and they work, when they work, because something concrete changed in the gap.

So stop asking how long. Ask what it’s for, what the rules are, when it ends, how you’ll decide, and above all what will be different when it’s over. If you can’t answer that last one, the length of the break is the least of your problems, because you’ll be coming back to exactly the relationship you left.

Three related questions come up constantly alongside this one. Taking a break covers the version that works and the several that do not. Taking space is the smaller, healthier cousin of a formal break and worth distinguishing. And if the break turns out to be an ending, healing after a breakup is where to go next.

Frequently asked questions

Sources

1. Dailey RM, Pfiester A, Jin B, Beck G, Clark G. (2009). On-again/off-again dating relationships: How are they different from other dating relationships? Personal Relationships, 16(1), 23–47. (On-off partners report greater conflict, more uncertainty about the relationship’s future, and lower satisfaction and commitment.)
2. Halpern-Meekin S, Manning WD, Giordano PC, Longmore MA. (2013). Relationship churning in emerging adulthood: On/off relationships and sex with an ex. Journal of Adolescent Research, 28(2), 166–188. (Nearly half of young adult daters and cohabitors report a breakup followed by reunion.)
3. Halpern-Meekin S, Manning WD, Giordano PC, Longmore MA. (2013). Relationship churning, physical violence, and verbal abuse in young adult relationships. Journal of Marriage and Family. (Churning associated with higher rates of physical violence and verbal abuse; association at population level.)
4. Vennum A, et al. (2014). Relationship cycling and outcomes. (Partners in cyclical relationships report lower explicit decision making, commitment and satisfaction, with greater conflict and uncertainty than non-cyclical partners; roughly 30–50% of young adult dating partners report at least one breakup and reconciliation with their current partner.)
5. Halpern-Meekin S. On the importance of “deciding, not sliding”: evaluating whether the factors that caused a breakup have genuinely been addressed, rather than reuniting to relieve the pain of separation.
6. Note on duration: no published research establishes an optimal length for a relationship break. Any specific figure offered elsewhere should be treated as opinion, including the reasoned default suggested in this article, which is presented as judgment rather than evidence.