Why does a couple's agreed ritual keep not happening, and how do you fix it?

What each one is actually for

Secure Functioning

Secure functioning treats a relationship as something the two of you build rather than something you have. Tatkin's model asks partners to agree explicitly that they operate as a two-person unit, protecting each other ahead of everything that competes for their time and loyalty: work, phones, extended family, even, in specific bounded moments, their children. Tatkin calls the result a "couple bubble." A large part of the practical work is deciding which small repeatable acts hold that up — how you say goodbye, how you come back, what happens in the ten minutes after one of you has had a terrible day — and treating those acts as commitments rather than as things you do when the mood is right. There is a second commitment in the model that matters as much and gets less attention: partners agree to stop treating each other's lapses as evidence about character. The alternative to that agreement is a relationship where every missed moment becomes a small piece of testimony in an ongoing case about how much someone cares.

Behavior Design

Behavior design starts from a claim that sounds cynical and mostly is not: what people do is predicted much better by how easy it was than by how much they wanted it. So the work happens on the surroundings, not on the willpower. Three tools do most of the labour. Cues: putting a visible, unmistakable trigger in the path you already walk, so that starting does not depend on remembering. Friction: counting the actual physical steps between where you are and the first second of the behaviour, then deleting some of them for things you want and adding some for things you do not — the running shoes by the door, the phone charging in the kitchen rather than the bedroom. Defaults: setting what happens when nobody makes a decision, since that is what happens most of the time. The move is deliberately unheroic. You are not trying to become the kind of person who does the thing. You are arranging matters so that the thing is what happens when you are tired and no one is being impressive.

What you get out of pairing them

The reason to care about behavior design is that it takes a whole category of struggle off the table. When the environment is arranged well, you stop needing to be in good condition for the right thing to happen, which matters because the days that decide a relationship are almost never the days you are in good condition. And what secure functioning offers is subtler and, for a lot of couples, the more valuable half: it removes the interpretation. Once you have both agreed what protects the relationship and both agreed to stop reading misses as evidence about each other, a skipped evening becomes a broken mechanism rather than a broken promise, and a broken mechanism can be fixed on a Tuesday without anybody having to defend their character. Hold both and something specific becomes available — the ability to say, out loud and without an argument starting, "our check-in isn't happening; what's in the way?" That sentence is almost impossible to say in a household where the missed check-in has been quietly accumulating meaning for a month. It is easy in a household where both people already agreed that the design, not the person, is the first suspect.

What Secure Functioning rests on

• Attunement requires active monitoring not good intentions clinical practice
  Staying attuned to a partner's internal state requires ongoing structured attention rather than goodwill alone because connection erodes by default without it.
• Explicit commitment changes daily priority weighting clinical practice
  Explicitly and mutually committing to protect the relationship above other priorities changes how partners weigh choices during ordinary daily conflict.
• Rituals of connection reduce reliance on spontaneous attunement practitioner report
  Building repeatable rituals of connection reduces dependence on spontaneously noticing a partner's needs in the moment.

What Behavior Design rests on

• Behavioral activation precedes motivation randomised trials
  Taking action before motivation appears, rather than waiting for motivation to arise first, itself generates the motivation and improved mood that waiting would delay.
• Default effect shapes choice randomised trials
  People disproportionately stick with whatever option is set as the default so redesigning defaults changes downstream behavior without requiring willpower.
• Friction determines behavior plausible mechanism
  Reducing the physical or digital friction required to perform a desired action increases how often it occurs independent of motivation level.

Where they actually meet

A ritual of connection is already a piece of environment design aimed at a relationship, built by people who had never heard the phrase. Instead of trusting that each partner will feel like turning towards the other on a given night, you install a repeatable structure that runs whether or not either of you is in the mood — which is exactly the logic of putting the running shoes by the door. Secure functioning's own principle says it: repeatable rituals reduce a couple's dependence on spontaneously noticing each other's needs. Behavior design says the general version of the same thing and then hands over the tools the couples lineage never developed. There is a division of labour here that is unusually clean, and it goes in both directions rather than one. Secure functioning can say what has to happen reliably inside a couple's ordinary week — protect the reunion, keep the ten minutes after a bad day, do not let a third party into a disagreement — but it offers very little for making a specific ritual survive a tired Tuesday, and it does not distinguish between a ritual that failed because someone withdrew and one that failed because the kitchen is where the phones charge. Behavior design has that diagnostic precision and no standing whatsoever to decide what a particular couple should be protecting. What flips when you hold them together is what a missed evening means. In the old frame the miss is information about a person, and information about a person is very hard to take back. In the new frame the miss is information about a system, and a system is something two people can look at from the same side of the table.

The principle underneath both

Character explains far less of what a couple actually does on a Tuesday than the shape of their evening does. Both traditions are, underneath, arguments against the same reflex — the reflex to explain a behaviour by what kind of person produced it. Behavior design makes that argument from outside: the same person does the thing at a much higher rate when the thing is easier, without any change in how much they wanted it, which means wanting was never carrying the load you assumed. Secure functioning makes the same argument from inside a relationship and for a different reason. Tatkin's model asks partners to stop drawing character conclusions from lapses not because character is irrelevant but because those conclusions are usually wrong and always expensive: once one person has decided the other is someone who does not prioritise them, every subsequent piece of evidence gets read through it. Neither lineage owns the pattern. It is a claim about where behaviour actually comes from, and its practical consequence is the same in both fields — when something you both wanted did not happen, look at the arrangement before you look at the person.

What that makes the real goal

If arrangement is doing most of the deciding, the goal follows directly: make the thing you both want the easiest thing that could happen in that room at that hour, and make it easy for both people rather than for whoever is more organised. State it that way and it stops being about couples or about behavioural science. It becomes a question you could ask of a team's stand-up, a family's dinner, or your own morning — is the wanted thing currently the path of least resistance, and if not, what exactly is standing in it? Two consequences fall out that neither tradition's own vocabulary makes obvious. The first is that the objective is measurable without anyone reporting a feeling: count how many steps stand between the cue and the first second of the behaviour, and count how often it happened. The second is that it is symmetrical. An arrangement that is frictionless for one person and effortful for the other is not solved; it has simply moved the whole cost onto one side, and in a household that cost gets paid in resentment rather than in missed evenings.

Other practices that serve the same goal

Other practices reach the same objective by different roads. Choice architecture, the applied version of default-setting studied in behavioural economics, has strong randomised and field evidence — organ-donation defaults, retirement-savings enrolment — and its transferable lesson for a household is that the option requiring no decision wins far more often than anyone predicts. Habit stacking anchors a wanted behaviour to one that already reliably happens, so that an existing routine supplies the trigger; the evidence is thinner than the choice-architecture literature but it solves the practical problem of finding a cue that actually fires. Commitment devices — locking the option away rather than making it harder, like a timed box for phones — have field-experiment support where the temptation is strong and predictable, and they suit exactly the couples who have already tried the gentle version twice. Behavioral activation, a well-evidenced treatment for depression, pursues the objective from the mood side: schedule the action first and let the wanting arrive afterwards, which has randomised-trial support in clinical populations and is the direct answer to "we'll do it when we feel closer." And Gottman's rituals of connection, from a couples research tradition separate from Tatkin's, arrive at the same place with observational evidence and one real advantage — they were written for two people from the start, where most of the design literature assumes a single actor.

A tradition that already does this

Weekly sabbath practice has been doing this for a very long time, in a form no design textbook would improve on. In observant Jewish households, Shabbat removes the competing options structurally rather than asking anyone to resist them — work stops, devices go off, the meal is at a fixed hour, the table is set the same way — and the connection that follows is not a matter of anybody remembering to prioritise the family that evening. It is what is left once everything else has been switched off. That is an environment redesign for the sake of relationship, arrived at religiously rather than behaviourally, so anyone uneasy that engineering an evening is a modern productivity habit being smuggled into a marriage should sit with that for a moment. Traditions have been building defaults for connection for centuries; what the behavioural literature adds is an explanation of why removing the option works better than resolving to ignore it.

Using Secure Functioning (PACT) to enhance Behavior Design

Behavior design is powerful and completely silent on what deserves the effort. It will make anything easier with equal enthusiasm, which is a problem in a household where the list of possible improvements is infinite and the appetite for being optimised is not. Secure functioning supplies the selection criterion, and it is narrower than people expect: engineer only the acts the two of you have explicitly agreed protect the relationship. Sit down and name three — realistically it is three, not ten — and let everything else stay unengineered. A useful test for whether something belongs on the list: would its absence be felt by the other person as being dropped? The reunion at the door usually passes that test. Loading the dishwasher does not, however irritating it is. The second contribution is structural and it is the one that saves people from doing real damage. In a shared home, a default is not yours to set. Moving the charger, declaring the bedroom device-free, putting a basket by the door — each of these looks like a design tweak and functions as a rule change imposed on someone who was not asked. PACT's insistence that agreements are made out loud and mutually is the correction: propose the change, name what it costs the other person, and accept that a change one of you cannot live with does not get made even if it would obviously work. The third contribution is the diagnostic order. Agree in advance, while nothing is wrong, that when a ritual stops happening you look at the arrangement before you look at each other. Write it down if you can. That agreement is what buys you the ability to raise a missed check-in as a logistics question rather than as an accusation, and it is worth more than any single design change you will make.

Using Behavior Design to enhance Secure Functioning (PACT)

When a couple's agreement keeps not happening, the story that forms is almost always about the other person. The check-in was agreed a month ago and has occurred twice, and by week three somebody has privately concluded that the other one does not really care — which is precisely the verdict PACT is built to keep partners from handing each other. Behavior design gives you something concrete to do before that verdict sets, and it takes about twenty minutes. Run a friction audit on the ritual, together, out loud. Ask four questions and answer them literally rather than charitably. What is the cue, and can both of you perceive it from where you actually are at that hour, or is the cue really "whoever thinks of it first"? How many physical steps stand between the cue and the first second of the ritual — finding each other, clearing the table, moving the laundry off the chair, putting down two phones? What is the default if nobody decides anything, and be honest, because for most households the honest answer is "we both look at our phones on the sofa." And who has to initiate, because a ritual that always needs the same person to start it is not a shared ritual, it is a service one of you provides. Then change the environment rather than the intentions. Pick a physical cue that arrives on its own — the dishwasher starting, the last bedroom light going off. Cut two steps: keep the chairs clear, put a basket by the door that phones go into on the way past. Set the default: the phones are in the basket, so remaining on the sofa scrolling is the option that now takes effort. And put the initiation on an object rather than a person, so neither of you is the one who always has to ask. There is a second move behind all of this, and it comes from the finding that action tends to come before motivation rather than after it. Do the check-in on the nights you do not feel close, deliberately, as the design rather than the exception — because "we'll do this when we feel more connected" makes the feeling a prerequisite for the thing most likely to produce it.

The nightly ten minutes you agreed to, that has happened twice in three weeks

You made the agreement on a good night — on the couch after dinner, feeling close, both of you meaning it entirely. Ten minutes, every evening, no phones, just how are you actually. Three weeks later it has happened twice, and something has started forming that you have not said out loud: that your partner does not protect this the way they said they would. Notice that the story is already doing work. You have begun noticing supporting evidence, which is what a story like that does, and you have not yet noticed a single piece of evidence about the room. So look at the room instead, and be unflatteringly literal about it. Dinner ends at a different time every night depending on who cooked, so there is no hour that reliably arrives. Both phones are already in your hands from the pickup-logistics thread, which means the ritual starts from a position of two people already engaged elsewhere. The sofa has laundry on one end. And there is no cue at all — the plan quietly assumed that whoever remembers first will say something, which on a tired Wednesday means neither of you, and on a good Wednesday means the same person every time. Not one of those four facts is about how much anybody cares. Now redesign it, together, in one short conversation. The dishwasher starting is the cue, because it happens every night, makes a noise, and neither of you controls when it lands. Phones go in a drawer in the kitchen at that exact moment — not face-down, in a drawer, because a visible phone keeps a slice of both your attentions on standby. Two specific chairs stay clear, and the laundry lives somewhere else. Nothing about the commitment changed; what changed is that it no longer requires either of you to be the one who remembers on a bad night. And here is the part most couples find surprising when they try it. On the first few evenings you will not feel like doing it, and you should do it anyway — that is the design working rather than failing. What tends to happen is that the wanting shows up three or four minutes in, once you are already sitting down and the day has started coming out. It does not show up beforehand, on the sofa, while you are deciding. That is why waiting to feel close before you talk almost never gets a couple to the talking.

When this is worth reaching for

Three weeks in, the agreement has been honoured twice, and one of you has quietly started building a theory about the other. That is the moment this matters most, because the theory is cheap to dismantle now and very expensive later. Some other conditions point the same way. A ritual that only ever happens when one specific person initiates it — that is a design fault presenting as a generosity imbalance, and it curdles. A stretch where capacity is genuinely low, because a newborn, a bad quarter or an ill parent is exactly when a system has to run without anybody being on form. A house that has recently changed — a move, a new job, a child starting school — since the old cues quietly stopped firing and nobody noticed which ones. There is also a clear not-yet. If the agreement itself is only a week old and has simply not had time, leave it alone; you have nothing to audit, and rebuilding a ritual that has not yet had a fair run tends to produce a more elaborate version of something you were never actually doing.

Where this does not apply

When what is missing is not a ritual but trust or agreement itself — a real rupture, a betrayal, or one partner who never actually said yes to the commitment being engineered around — friction reduction has nothing to work on, because making a behaviour easier only raises the rate of something already wanted. It cannot manufacture the wanting. Running the design conversation on top of an unaddressed rupture also has a specific cost: it makes the household look like it is functioning while the thing underneath goes unnamed for another month. There is a second limit that people who like systems tend to walk into anyway. Redesigning a shared environment is only legitimate once both partners have explicitly agreed to what is being engineered for. Absent that agreement, changing a household default unilaterally is not practising either method — it is one person rearranging the house and calling it a system, and the other person will experience it exactly as that. Watch for the tell, which is that you find yourself explaining the new arrangement rather than having agreed it. And a smaller one worth naming: a ritual can be perfectly engineered and completely hollow. If the phones are in the drawer, the chairs are clear, the cue fires every night and neither of you is actually saying anything true in those ten minutes, more design is the wrong response. The mechanism is running; the content is what needs attention.

What the evidence does and does not support

Take the two halves separately, because they are not close in strength. The default effect — that people disproportionately take whichever option requires no decision — is among the better-replicated findings in behavioural science, with randomised and large-scale field evidence behind it, though most of that evidence comes from single decisions like enrolment and donation rather than from a behaviour repeated nightly by two people. That the wanting often follows the doing rather than preceding it carries randomised-trial support from behavioral activation in the treatment of depression, which is a clinical population and a specific protocol, not a general licence to assume every reluctance dissolves once you start. The friction claim is graded lower on purpose: it is mechanistic, a well-motivated account of a pathway rather than a directly measured outcome, and it is the one most often stated with more confidence than it has earned. On the other side, what this model says about monitoring, and about a stated commitment changing how partners weigh an ordinary Tuesday, comes from clinical observation in Stan Tatkin's couples practice rather than from controlled trials, and the value of connection rituals specifically rests on practitioner report. Nobody has tested the combination. The study that would settle it is straightforward to describe and has not been run: couples who have agreed the same ritual, randomised to receive a design audit or not, measured on how often the ritual actually occurred over three months — occurrence, not satisfaction, because satisfaction is exactly where hopeful reporting creeps in. What you can do meanwhile is run the smaller version on yourselves: count the occurrences for three weeks before you change anything, and three weeks after.