By 
Friends drift apart for two different reasons, and most advice treats them as one. Sometimes the setting that held the friendship up went away, and the two of you still fit. Sometimes you both changed, and the fit itself is gone. The first needs a small, regular place to meet. The second needs something harder: downgrading the friendship on purpose and letting yourself grieve it. We call this context loss versus divergence. It's our framework, a synthesis of published research and our own observation, and we'll mark which parts are which.
Why the usual advice stalls
Search for growing apart from friends and the pages agree on three things. It's normal, here are the signs (fewer texts, forced conversation, plans that slide), and you should either reach out or let go. Each piece of advice is right for one kind of drift and wrong for the other. Reaching out to someone you've outgrown gets you a stiff coffee and a month of guilt. Letting go of someone you simply stopped seeing throws away a friendship that a standing plan would have saved. The missing step is working out which one you have.
What the research does show
The Survey Center on American Life asked 2,019 U.S. adults in May 2021 about their friendships. In the previous 12 months, 47 percent had lost touch with at least a few friends and 9 percent with most of them. That was mid-pandemic, so read it as a high-water mark. The same survey asked where people with close friends had met them (more than one answer allowed): work 54 percent, school 47 percent, an existing friend network 40 percent, the neighborhood 35 percent. Most of those are places that put you in the same room as the same people again and again.
That detail connects to Gerald Mollenhorst, Beate Volker and Henk Flap (2014, Social Networks), who followed Dutch personal networks for seven years. The average number of confidants barely moved, but individual relationships turned over a lot, and the authors found that a lack of chances to meet, across work, family, clubs and the neighborhood, was a major reason relationships ended. New ties also tended to form in settings where people already met members of their network. Setting matters more than most of us assume.
Two more studies set the scale. Jeffrey Hall (2018, Journal of Social and Personal Relationships) surveyed 355 recent movers and followed 112 college freshmen, and estimated about 50 hours together to move from acquaintance to casual friend, about 90 to friend, and more than 200 to close friend. Time at work counted for less than time spent on leisure. And Sam Roberts and Robin Dunbar (2015, Human Nature) tracked whole personal networks for 18 months through a major life transition. Friendships tended to lose quality, and extra effort partly offset the loss: for women, talking more often, and for men, doing more things together.
None of these studies sorted drift into two kinds. That part is ours, and the box near the end spells out the limits.
Two kinds of drift
Lost context. The friendship was held up by a shared setting: the same office, dorm, street, team, or school pickup line. The setting ended, and nothing replaced it. Neither of you changed much. The signature is warmth with long gaps: you're glad every time you meet, and you meet rarely, and one of you says "we should do this more" at the end of each one. Our piece on work friendships covers the most common version of this.
Divergence. The people changed. Values, pace of life, what each of you wants from a friend. One of you had a baby, got sober, got promoted, left a marriage, or lost interest in the thing you used to do together. Sometimes it's money, when one of you can afford the trips and dinners and the other can't (the real price data is in our friendflation piece). The signature is effort even when time isn't the problem. You can clear a Saturday and it still feels like work.
The four-question test
Answer these about one specific friend, and answer fast.
- The new-setting question. If you met this person today, in a class or at a neighbor's party, would you want a second coffee?
- The ten-minute question. When you do talk, does it get easy within ten minutes, or are you catching up like two people reading from a script?
- The missing-thing question. If a free Saturday appeared, would you want to spend it with them? Is the thing missing time, or is it fit?
- The initiation question. Before the setting changed, who started plans, and was it lopsided then too?
Yes to the first three points to context loss. No to the first three points to divergence. Mixed answers usually mean you're somewhere between, which is common. The fourth question isn't scored. If the pattern of who reached out was lopsided before the move or the job change, the setting didn't cause it, and you're in the third case below.
If it's lost context: build a small one
You don't need a heart-to-heart. You need a recurring reason to be in the same place. Choose one slot that costs almost nothing to keep: dinner the first Thursday of the month, a Saturday walk, a standing game night. Say it as a plan and not a wish: "First Saturday, 9 a.m., the park loop. Want in?" Hall's numbers suggest realistic expectations, though they describe forming friendships, not maintaining them. A two-hour dinner every month is about 24 hours a year, a fraction of the 200 hours or so for a close friendship, so this keeps the friendship alive and doesn't rebuild closeness fast. Give it three or four meetings before you judge it. If you've been out of touch for a long stretch, our reconnecting guide covers the first message.
If it's divergence: downgrade on purpose, and grieve
No schedule fixes a fit problem, and trying to can make you resent the friend. The alternative isn't a dramatic exit. Move them to a smaller place on purpose: a birthday text, a yearly lunch, a warm reply when they post good news. Stop measuring the friendship against what it was. Then let the loss count. People tend to treat a faded friendship as too small to mourn, which is why grieving a friend is its own skill. Putting freed-up time into fewer, closer people is the idea behind slow friendship. If what you're seeing goes beyond mismatch into being drained or belittled, read toxic friends instead, because that's a different problem.
The case most people live in: one-sided drift
Often only one person has moved. You changed and they didn't, or they changed and you didn't, or you realize you're the one who stopped initiating. This is the most uncomfortable version, because the other person may still be fully in the friendship. We don't think it helps to assign blame. Changing is not a betrayal, and neither is being the one who stayed the same.
Two practical moves. First, check who's carrying the effort. Our better-friend piece looks at who does the reaching out and how reaching out gets treated, both easy to audit. Second, ask once, plainly: "I've noticed we don't talk much, and I've missed you. Want to get something on the calendar?" Whatever comes back is information. A warm yes means context loss was part of it. A vague answer, repeated, means you have your answer, and it's still not a verdict on either of you. If you're also finding it hard to replace people, the reasons are in why it's hard to make friends after 30.
What this is based on, and what it is not. The two-cause framework and the four questions are ours. They're a synthesis, they haven't been tested, and we have no member data on them. What the cited research supports: friendships often end where meeting chances end (Mollenhorst and colleagues), time together builds closeness (Hall), effort offsets decay (Roberts and Dunbar), and losing touch is common (Survey Center). What it doesn't show: that drift sorts cleanly into two kinds, that our test predicts anything, or that a standing plan rescues a friendship. Hall's samples were recent movers and freshmen, the Survey Center figures come from one pandemic-era survey, and we read the Hall paper through its university summary and the Roberts and Dunbar paper through its abstract. We used no NexSpark member data.
Common questions
What are the signs you're growing apart from a friend?
Common signs are conversations that take effort in the first ten minutes, plans that keep sliding, and a small feeling of relief when one gets canceled. Those signs show up in both kinds of drift, so they can't tell you which one you have. The four-question test on this page can.
Why do friends grow apart?
In our framework, for one of two reasons: the shared setting went away (lost context) or the two people changed in ways that no longer fit (divergence). A 2014 study of Dutch personal networks over seven years found that a lack of chances to meet, across work, clubs, family and the neighborhood, was a major reason relationships ended. We found no study that measures divergence directly.
Is it normal to grow apart from friends?
Yes. In the Survey Center on American Life's May 2021 survey of 2,019 U.S. adults, 47 percent had lost touch with at least a few friends in the previous 12 months and 9 percent with most of them. That was during the pandemic, so it may overstate an ordinary year. Normal doesn't mean there's nothing to do about it.
Should I say something to a friend I'm drifting from?
It depends on the kind. If the cause is lost context, skip the big talk and propose a specific plan. If it's divergence, a short, kind note works better than silence: you've missed them, you think you're in different places, and you'd still like to hear from them now and then. If you suspect the drift is one-sided, ask once and take the answer.
How do I save a fading friendship?
Start by finding out whether it's context loss, then set one standing plan, such as a monthly dinner or a walk on the first Saturday. Be realistic about the math: Jeffrey Hall's 2018 estimate for a close friendship is more than 200 hours, and a two-hour monthly dinner adds about 24 hours a year. A standing plan keeps a friendship alive, but it won't quickly deepen one.
Sources
- Daniel A. Cox, The State of American Friendship: Change, Challenges, and Loss, Survey Center on American Life, June 2021 (2,019 U.S. adults, May 14 to 23, 2021, margin of error 2.4 points).
- Gerald Mollenhorst, Beate Volker and Henk Flap, Changes in personal relationships: How social contexts affect the emergence and discontinuation of relationships, Social Networks 37, 2014, pp. 65 to 80.
- Jeffrey A. Hall, How many hours does it take to make a friend?, Journal of Social and Personal Relationships, 2018 (DOI; summarized by the University of Kansas).
- Sam G. B. Roberts and Robin I. M. Dunbar, Managing Relationship Decay: Network, Gender, and Contextual Effects, Human Nature 26(4), 2015, pp. 426 to 450.