The Relationship Check-In That Does Not Turn Into a Fight

A relationship check-in is a short conversation you have on a schedule, about how the two of you are doing. This one is five questions and twenty minutes, in the same slot every week.

Couples rarely stop because the questions were bad. They stop because week two ran to an hour, or week three landed on the evening of an argument, or somebody answered honestly and the other person explained why they were wrong to feel that way. After that, nobody suggests it again.

So most of what follows is mechanics. When to hold it, who answers first, what to do when the twenty minutes are up and you are not finished, and what to say when an answer stings. The five questions are in the middle.

A check-in you run badly for a year beats a check-in you run beautifully for three weeks.

Pick the slot before you pick the questions

The check-in that never happens is the one you agreed to have whenever things feel off. Nothing in that sentence tells either of you when to start.

Peter Gollwitzer and Paschal Sheeran have spent decades on the gap between meaning to do a thing and doing it. A goal intention is “I intend to achieve X.” An implementation intention, in their own summary, “only spells out the when, where, and how of what one will do.” Across the 94 studies in their 2006 review, forming that kind of if-then plan had a medium to large effect on whether people reached the goal (d = .65), and a similar effect on the specific failure of never getting started (d = .61).

None of that is couples research. It is mammograms, vitamin pills and exercise. The gap it measures is the one that kills check-ins, which is the distance between two people agreeing something is a good idea and either of them beginning it.

So name it. Sunday at five, at the kitchen table, before dinner. Put it in both calendars as a repeating event with a title on it.

A fixed slot also does something a good intention cannot. If the check-in only happens when one of you asks for it, then asking becomes a signal that something is wrong, and the check-in turns into a summons. The weeks when neither of you has anything to report are what make the week you do have something survivable.

Hold it somewhere you can walk away from. The last twenty minutes before sleep is the worst slot available, because neither of you can end the conversation by leaving the room.

Five questions, and both of you answer all five

In 1997 Arthur Aron's team put pairs of students who had never met into a room for 45 minutes with three envelopes of question slips, fifteen minutes to an envelope, the questions getting more personal within each set and across the three. One person read a slip aloud, both answered it, and they alternated who read the next one.

Those pairs rated themselves closer afterwards than pairs given small talk, 4.06 against 3.25 on the same measure, a gap the authors put at .88 standard deviations. Their own conclusion about the design is the reason to choose your questions rather than sit down and see what comes up: the result “is not simply a matter of putting two people together in any kind of structured interaction for 45 min.”

Those were strangers, so it is evidence about how closeness gets built rather than about your marriage. Two pieces of the design are worth taking anyway. The questions escalate, and the two of you alternate who goes first.

1. What did I do this week that you want me to do again?

   This one has to be specific or it does nothing. “You have been great” is not an answer. “You took the bins out on Tuesday without me asking” is.

   Asking for a repeat is what separates it from a compliment. You finish the question holding one concrete thing to do next week.

   If neither of you can name anything, sit in that for a second. The check-in has done something useful inside the first two minutes.

2. When did you feel furthest away from me this week?

   This is the one that does the most work, and it is worth saying so rather than pretending the five are equal. It is the only question here whose honest answer is usually about the person who asked it.

   “When” is deliberate. A moment can be answered. “Why do you feel distant from me” asks your partner to diagnose the relationship on the spot, and almost nobody can.

   It catches things while they are still small. A Thursday evening when you were on your phone is an easy thing to hear about. The same information six months later arrives as a pattern, and patterns are much harder to hear.

3. What is on your plate right now that I am not seeing?

   The logistics question, and it is here so the check-in does not become entirely about feelings. Renewals, a work thing that has been sitting there two weeks, a call to a parent, the appointment somebody keeps meaning to book.

   Noticing what needs doing is work whether or not either of you counts it as work, and it lands unevenly in most households; we wrote about where that cost falls hardest. Answering out loud moves an item out of one head and into two.

4. What is coming up that you are dreading?

   This points the conversation at the week ahead, and dread is usually specific: a meeting, a phone call, a visit, a scan, a Saturday that is going to be long.

   You are not required to fix any of it. Knowing it is there is most of the value, and the other person stops guessing why you were short with them on Wednesday.

5. What is one thing I could do next week that you would notice?

   It ends on an action instead of a feeling, which is what makes twenty minutes a ritual rather than a debrief.

   “That you would notice” is carrying the question. It rules out the effort you make privately, which is real effort and is also invisible to the person it was for.

   One thing each, not a list. The Gottman Institute's weekly meeting closes the same way, with each partner naming one thing the other could do in the coming week to help them feel connected.

Who answers first, and why the first three minutes decide the rest

In 1999 Sybil Carrere and John Gottman published a study of 124 newlywed couples in Family Process. Each couple talked through a real ongoing disagreement for fifteen minutes in a lab. The tape was coded for emotion in five three-minute blocks, and the couples were followed for six years, over which seventeen of them divorced.

The first three minutes predicted the outcome. For the wives, those three minutes predicted as powerfully as the remaining twelve did; for the husbands the gap between the two groups widened as the conversation went on.

That is a prediction and not a cause. What it says is that by three minutes in, the shape of the conversation is already visible, and a check-in is only twenty minutes long.

Two rules come out of it. Alternate who answers first, question by question, so neither of you is permanently the one who sets the tone and neither is permanently the one who follows. And if you have been sitting on something all week, answer second on that question.

The paper also draws a distinction worth memorising. Describing a wife opening a conflict discussion, Carrere and Gottman write that a criticism implies “a global issue reflecting something defective in the husband's character that has caused the problem, whereas the complaint is more specific and does not suggest that the problem is with the husband's defective personality.”

Swap the pronouns and that is the whole skill. “You never think about my schedule” is a verdict on a person. “I found out about Saturday on Friday night, and I had already told someone I was free” is a Friday night, and a Friday night can be answered.

Twenty minutes, and what to do when it runs over

Set a timer and leave the phone face up on the table doing nothing else. Twenty minutes is five questions and two people, so roughly two minutes an answer.

It is short on purpose. The Gottman Institute recommends an hour a week for its own version of this meeting, and an hour is a better conversation than twenty minutes. It is also the reason a lot of couples never hold the first one.

When the timer goes and you are not finished, stop and book the rest. A named time in the next two days, not “later this week,” which means no.

Nearly every overrun is one topic, and one topic deserves a slot of its own, where neither of you is watching a clock.

Nothing raised in a check-in has to be solved in the check-in. Naming it is the entire job, and that is the rule that keeps twenty minutes survivable.

When an answer stings

It will. You asked a question designed to surface something and it surfaced. What happens in the four seconds afterwards decides whether there is a check-in next week.

The move is to ask one more question instead of answering. “When else?” or “Say more.” Your first instinct will be to supply context, and context sounds like a defense even when every word of it is true.

You are allowed to say you need to think. “I do not have a good answer for that tonight. Can I come back to it on Sunday?” is a real answer, and it is better than the one you will produce while stung.

The Gottman Institute gives its own weekly meeting two instructions for this, both worth borrowing word for word if you find yourself sharpening: non-defensive listening, and a softened start-up in the form “I feel [emotion] about what [situation]. I need [positive request].”

Quoting an answer back during an argument three weeks later will end the check-ins for good. Once is enough to do it.

If one of you goes quiet or cries, finish the question you are on, then decide together whether to keep going or stop at four.

Five ways it falls apart

1. It becomes a list of grievances

   Question one is not optional and it goes first. A check-in that opens on what went wrong is an argument with a timer attached.

2. You hold the first one after a fight

   A check-in is maintenance. Holding your first one on the evening of a bad argument teaches both of you that this is where the difficult conversations live, and neither of you will look forward to the second. Repair the argument on its own terms and keep the scheduled slot.

3. One of you prepares and the other does not

   The Gottman version uses notebooks. This one does not, on purpose. If one of you arrives with three written points and the other arrives from the kitchen, the second person is behind before the first question, and the check-in has become one person's meeting.

4. It turns into the household admin meeting

   Question three is the only one that gets to be about logistics. Anything else it raises goes on the list you already keep.

5. It gets moved to sometime this week

   It can be moved, but only to a named time. A check-in rescheduled to “this week” has been cancelled by two people who would both deny cancelling it.

Where we come into this

Twenty minutes a week is maintenance. It will not do the job that spending real time together does, and it is not meant to.

What stops most couples going out is the arranging. One of you picks the thing, finds the evening, books it, and then carries the whole plan around until Friday, every single time. We take that part off you: you answer some questions about the two of you, and a full date plan comes back, with the order of the night, the small personal touches, and what to do when the restaurant loses your booking.

One month of Big Moments Premium is free at the moment, and it does not ask you for a card.

Claim your free month of Big Moments PremiumJust your email. No card, no commitment.

If you want to run one this week

Pick the slot now, while you are still reading, and put it in both calendars with a name on it.

Write the five questions somewhere you will both see them. Set a timer for twenty minutes when you sit down, start with question one, and swap who answers first as you go.

If it runs over, book the rest before either of you stands up.

Research referenced

• Carrere, S., and Gottman, J. M., “Predicting Divorce among Newlyweds from the First Three Minutes of a Marital Conflict Discussion,” Family Process, 1999.
• Gollwitzer, P. M., and Sheeran, P., “Implementation Intentions,” and the 2006 meta-analysis it summarises, in Advances in Experimental Social Psychology.
• Aron, A., Melinat, E., Aron, E. N., Vallone, R. D., and Bator, R. J., “The Experimental Generation of Interpersonal Closeness,” Personality and Social Psychology Bulletin, 1997.
• The Gottman Institute on its State of the Union meeting: how to hold one and how to start one. Cited for what the institute recommends, not as research.