Why couples drift apart

What the research supports, and what it does not

This is a field where popular ideas outrun evidence, so it is worth separating the two carefully.

Reasonably supported: that everyday responsiveness matters more than grand gestures; that how a couple handles conflict predicts more than whether they have it; that repair after a rupture matters more than avoiding ruptures.

Not supported, despite being everywhere: the love-languages framework. Impett, Park & Muise (2024), in Current Directions in Psychological Science 33(2):87–92, examined its three core assumptions and found none well supported. People tend to rate all five expressions as meaningful rather than having one primary language. Factor analyses "deviated from the proposed five-love-languages structure." And there is "no empirical support that couples in which partners match report higher relationship satisfaction." The authors propose a "nutritionally balanced diet" model instead — that people need all of it, in varying amounts, rather than one channel.

Contested, and we will not use it: the frequently-quoted claim that marital outcomes can be predicted with 94 percent accuracy. Heyman & Slep (2001), in the Journal of Marriage and Family, argue that the accuracy figures came from models fitted and evaluated on the same sample without cross-validation — which measures fit rather than prediction. The statistic circulates constantly. It is not one we are willing to repeat. Our full write-up.

The failure mode of every relationship quiz

Here is the criticism that shows up across reviews of the main couples apps, in customers' own words: the quiz produces a result and then nothing happens.

You answer twenty questions. You are told you are a particular type. And then you are back where you started, holding a label with no instruction attached. Paired, Lasting and Relish all attract this complaint. It is the single most common thing users say about the category.

A category label is not an action. Knowing that your partner values a certain kind of attention does not tell you what to say at 9pm on a Tuesday when the room has gone quiet. That gap between insight and behaviour is where most of these products stop.

Nearvo's editorial rule follows directly from that: every explainer ends in a script or a ritual, not a category. If a page cannot end with something you could do this evening, it does not go up.

Four situations where drift accelerates

After a baby. Sleep is gone, evenings are gone, and every remaining conversational slot is logistics. The couple has not fallen out; they have simply stopped having any input that is not a handover. Twenty questions for five spare minutes.

When the same argument keeps recurring. Some disagreements are not solvable, and the published relationship-science position is that a substantial share of couple conflict is perpetual rather than resolvable. Treating a perpetual problem as a solvable one produces the same fight on a loop. Twenty-five questions for that.

Long distance, or long hours. Contact stops being spontaneous and becomes scheduled, and scheduled contact drifts toward reporting rather than connecting.

When everything is fine. The most common and least discussed. Nothing is wrong, so nothing gets attention, and the number of bids per day quietly falls for a year.

What actually helps

Notice more bids. This is the highest-leverage and least dramatic change available. You cannot respond to something you did not register. Fifty concrete examples, with responses.

Have a regular time to talk that is not triggered by a problem. A conversation that only happens when something is wrong teaches both people that the conversation means something is wrong. The weekly check-in agenda, and why the order matters.

Repair after conflict rather than trying to avoid it. The questions that work after a fight are ordered — understand first, repair second, prevent third — because doing them out of order restarts it. Twenty-five questions for after a fight.

Ask about something other than logistics. Most long-term couples know each other's schedule perfectly and have not asked a genuinely new question in a year.

Where Nearvo fits, and where it does not

Nearvo is a daily connection habit — one question a day, answered separately and then revealed to each other, with shared streaks that only advance when both of you show up, and decks built for particular moments rather than a generic pile of prompts. Planned pricing is $49.99/year, $9.99/month or $129.99 once, with one subscription unlocking both partners. It is in development and not released.

Everything on this site is free right now and works whether or not the app ever ships.

Questions and answers