What the data actually shows
The clearest articulation of this comes from Eli Finkel's work on what he calls the all-or-nothing marriage. His argument, drawing on historical and psychological evidence, is that our expectations of marriage have climbed toward self-expression and personal fulfilment — we now ask a spouse to help us become our best selves — while the time and energy many couples can invest has not risen to match. The result is that marriages can reach unusual heights when those expectations are met, but more often fall short of a standard that was once distributed across a whole community.
A separate line of research on social diversity points in a complementary direction. Studies of what is sometimes called relational or social-portfolio diversity find that people who engage with a wider variety of relationship types — partner, friends, family, neighbours, colleagues — tend to report higher wellbeing than those whose interactions are concentrated in fewer kinds of ties, even holding the total amount of social contact roughly constant.
Underlying both is a simple idea with decent support: different relationships are suited to meeting different needs. The person who is your steadying presence may not be the one who challenges you intellectually or shares a particular passion, and there is no strong reason to expect one person to be all of those things at once. Spreading needs across several relationships appears, on average, to be both more realistic and more sustaining.
Why this feels different from how it actually is
It feels like one person should be able to be everything partly because the cultural script says so. The dominant story about romantic love — in films, songs, and wedding language — is that the right person completes you and becomes your everything, which quietly sets one relationship up to carry a load that used to be shared.
Modern life also concentrates relationships in a way earlier generations did not. As extended family, dense neighbourhoods, and community institutions have thinned for many people, a partner can end up being the main or only close adult relationship by default — not because they are meant to meet every need, but because there are fewer other ties available to absorb them.
And when a strong relationship still leaves some need unmet, it is easy to read that as a flaw in the relationship or the person rather than as a normal limit of what any single tie can do. That misreading can put pressure on something that may, in fact, be working well.
One person can meet many of your needs, often very well — but expecting them to meet all of them tends to strain the relationship and the person.
What the research says to do about it
The direction most supported by this research is to build and maintain a portfolio of relationships rather than concentrating everything in one. Investing in friendships, family, and community alongside a partnership tends to be associated with higher wellbeing and, by spreading needs, can take pressure off the central relationship rather than competing with it.
Finkel's own framing suggests two broad routes when a relationship is carrying too much: either invest more time and attention to help it meet high expectations, or recalibrate expectations so the relationship is not asked to supply everything. Both are presented as legitimate; the trap is holding all-or-nothing expectations while supplying neither the investment nor the recalibration.
Matching needs to the relationships best suited to them is also sensible given that different ties meet different needs. Looking to a friend for a particular kind of companionship, or to a wider circle for intellectual or recreational connection, is not a failure of the primary relationship — it is closer to how the research suggests social needs are most sustainably met.
What the research says does not help
Treating any unmet need as evidence that you are with the wrong person tends to mislead, because no single relationship is likely to meet every need. Searching for someone who can be everything sets a standard the research suggests is rarely reachable, and can lead to discarding relationships that are actually strong.
Withdrawing from friends and outside connections to pour everything into one relationship tends to backfire. It removes the very diversity of ties that research associates with wellbeing and leaves the single relationship to carry more than it comfortably can, which raises the pressure on it rather than lowering it.
Holding sky-high expectations of a relationship while investing little time or attention in it is, in Finkel's framing, the least workable combination. High expectations are not the problem on their own — the mismatch between what is expected and what is invested is what tends to leave people disappointed.
Looking to a friend or a wider circle for a need your partner doesn't fill is not a failure of the relationship — it's how social needs are most sustainably met.
What this looks like in real life
The load one tie used to share
Confidant, best friend, intellectual equal, co-parent, source of passion and growth — these roles were once spread across family, friends, and community. When they all route through one person, that relationship carries a load it was never built to hold alone, and falling short of it can look like a flaw rather than a normal limit.
A strong relationship that still leaves a need unmet
Someone who is your steadying presence may not be the one who challenges you intellectually or shares a particular passion. Looking to a friend for that companionship, or to a wider circle for recreational connection, is not a failure of the primary relationship — it is closer to how the research suggests social needs are most sustainably met.
Real numbers in context
This is a topic where the evidence is more conceptual and historical than numerical, so it is worth being honest that there is no single headline statistic. The strongest claims are about pattern and direction: expectations of marriage have shifted toward self-fulfilment over time (Finkel's historical analysis), and greater diversity in one's mix of relationships is associated with higher wellbeing (the social-portfolio research).
Supporting context comes from population data on relationships more broadly — for example, that most adults report only a few genuinely close friends, which means the pool of ties available to share needs across is often smaller than people assume. Those figures vary by survey and country and should be read as rough indications rather than precise rates; the related insights on close friends and loneliness go into them further.