When researchers at Penn State and the University of Nebraska tracked the same married couples for 17 years, asking them at five separate points how happy they were, they expected the classic U-shape: happiness falling in the early years, bottoming out in the middle, then climbing back up as pressures ease. Jody VanLaningham, David R. Johnson, and Paul Amato found something less comforting. In their 2001 analysis in Social Forces, the U-shape only appeared when they compared couples of different marital durations at a single point in time. Once they followed the same couples across the actual years of their marriages, the rebound faded, and marital happiness tended to decline rather than recover on its own.
The “7-year itch” is not really about the number seven. It is about what happens when a relationship stops being new, and how consistently that shift shows up across large, tracked populations.
Where the number seven comes from
It comes from the 1955 Marilyn Monroe film of the same name, but demographers picked it up because it kept appearing in divorce records. Hill Kulu’s 2014 study in Demography, using Finnish marriage-register data spanning decades, modeled the risk of divorce by year of marriage and found a familiar pattern reported since the 1980s: divorce risk climbs in the first year, peaks around years five to seven, then gradually declines.
Kulu’s contribution ruled out a simpler explanation. A couple can sit in a low-satisfaction stretch for years without separating, which is where longitudinal satisfaction data becomes more useful than divorce statistics: it describes what is happening inside marriages that stay intact.
What the longitudinal data actually shows
The most complete picture comes from a 2021 meta-analysis in Psychological Bulletin by Janina Larissa Bühler, Samantha Krauss, and Ulrich Orth, who pooled data from 165 independent samples and more than 165,000 participants across 16 countries. They modeled two timelines at once: age and relationship duration, and the two produced different curves.
Plotted against age, satisfaction traced a U-shape, declining through early adulthood to a low point around age 40 before recovering. Plotted against relationship duration, the pattern was messier: satisfaction tended to decline across roughly the first decade, with the steepest average drop early on, before any later movement. The takeaway is that a dip in the early-to-mid years appears close to universal, and that time alone does not reliably repair it.
Read alongside VanLaningham’s panel study, the findings converge from different data: what tends to repair a dip is something couples do, not something that happens automatically. Justin Lavner and Thomas Bradbury’s 2010 study in the Journal of Marriage and Family, which followed newlywed couples over their first years, found meaningful variation in how satisfaction changed even within that early window. Couples who managed conflict effectively and stayed emotionally responsive in year one were significantly less likely to show the sharp declines that appeared in other couples by year three or four. The dip was not the differentiator; how a couple handled ordinary friction, before any dip appeared, predicted who avoided the steepest version of it.
Why the timing is so consistent
- Novelty wears off on a schedule. Early satisfaction is propped up partly by newness itself. As that supply runs out, so does one reliable source of positive feeling, independent of anything going wrong.
- The middle years concentrate stress. The years when satisfaction dips most sharply often overlap with young children, career demands, and financial strain arriving at once.
- Conflict patterns set early tend to hold. Lavner and Bradbury’s finding that first-year conflict style predicted later trajectories suggests the dip is not simply a random event.
The common mistake
The instinct, when satisfaction drops, is to read it as evidence the relationship itself was a mistake, that a good marriage should not feel this flat. The research points the other way. A dip is closer to the statistical default. Reading a normal decline as a verdict on the relationship’s viability is part of what can turn a predictable dip into a lasting one, because couples who interpret it that way tend to withdraw further, which removes the very responsiveness Lavner and Bradbury found protective.
Which couples this speaks to
Couples in the first decade of a relationship, especially those noticing a flatness they cannot quite name, are the ones this research speaks to most directly. If a marriage is in active crisis, ongoing conflict without resolution or one partner already checked out, the satisfaction-dip research explains less of what is happening, and a structured approach to rebuilding trust with professional support is more relevant than a timeline. For a lower-stakes but noticeable flatness, understanding that the timing is ordinary, and that what looks like disinterest is often closer to roommate syndrome than a real loss of feeling, can change how urgently a couple responds without changing whether they should.
How Relatio may help
Relatio’s relationship program is built around a premise the research tends to support: satisfaction often dips at predictable points, and predictable patterns may respond better to structured, consistent effort than to waiting them out. The program is designed to guide couples through daily exercises aimed at the mechanisms above, rebuilding novelty, supporting responsiveness during ordinary friction, and tracking what is changing week to week rather than relying on how one bad night felt.
Relatio is an educational wellness tool, not therapy, and is not a substitute for professional care. A licensed couples therapist may better serve couples in active crisis, and Relatio is designed to complement that kind of support rather than replace it.
FAQ
Is the 7-year itch a real, scientifically documented pattern?
The number seven specifically is closer to folklore than a finding. Still, the underlying pattern, a rise-then-fall in divorce risk and a comparable dip in reported satisfaction during the first decade of a relationship, is well documented, including in Hill Kulu’s 2014 Finnish register-data study and the Bühler, Krauss, and Orth 2021 meta-analysis.
Does marital satisfaction ever go back up after the dip?
It depends on which data you look at. Cross-sectional studies comparing couples at different ages often show an upturn later in life. VanLaningham, Johnson, and Amato’s panel study, which followed the same couples for 17 years, did not find that same rebound. The honest answer is that recovery seems to depend more on what a couple does than on the passage of time alone.
Is a satisfaction dip a sign that a marriage is failing?
Not by itself. The research treats a mid-relationship dip as close to the norm rather than the exception. What predicts real trouble, based on Lavner and Bradbury’s research on newlyweds, is less the dip itself than how a couple handles ordinary conflict and emotional responsiveness in the years leading up to it.
What actually helps during this stretch?
The strongest predictor in the newlywed research was consistent emotional responsiveness and effective conflict resolution, not any single grand gesture. Structured, regular effort, the kind built into evidence-based approaches to relationship maintenance, tends to outperform waiting for things to feel different on their own.
