Why Adult Female Friendships Can Be So Hard to Maintain

Most close relationships come with something that holds them in place. A marriage has a certificate and a shared address, a family has history and holidays, and a job has a paycheck and a calendar. Friendship has none of that scaffolding, so it runs entirely on two people choosing, again and again, to make room for each other. For adult women, that repeated choice competes with a growing list of other demands, and the research on why it gets harder is specific enough to act on.

Closeness is built from hours

Jeffrey Hall, a communication researcher at the University of Kansas, put numbers on something most people sense without measuring. In his 2018 study, it took roughly 50 hours together to move from acquaintance to casual friend, about 90 hours to reach the level of friend, and more than 200 hours before someone counted as a close friend. Hours spent working together counted for much less than other kinds of time. Those totals are easy to read past, so it helps to turn them into a calendar. A dinner of two hours once a month adds up to 24 hours a year, which means reaching 200 hours at that pace would take more than eight years. That arithmetic is mine, not Hall's, and ordinary life adds phone calls and group gatherings, but it shows why a friendship that receives only crumbs of time tends to stall at the casual stage.

Earlier in life, institutions supply those hours without anyone arranging them. School, dorms, and teams put the same people in the same room for hundreds of hours. Hall also points out that deeper connections tend to form during times of change, when people are meeting more new people than they normally would. Once a woman is settled into a job, a home, and a routine, every hour with a friend has to be arranged on purpose. That is a different kind of effort from the one that built friendships at nineteen, and most of us were never taught that it would be required.

Friendships fade unless someone keeps them going

Robin Dunbar, an Oxford University anthropologist, has spent decades studying how many relationships a person can hold and at what depth. His layers run roughly from five close friends to about 15 good friends, 50 friends, and 150 casual ones. The inner layers are small because closeness is expensive, and the price is time. Dunbar has written that friendships have a natural rate of decay when contact stops, that social media can slow the decay, and that occasional face to face contact is needed to keep a friendship from dying off. The important word is decay, because it describes a slow slide through the layers and not a break. Nobody has to do anything wrong for a close friend to become a friendly acquaintance.

One of Dunbar's studies suggests that women and men fight that slide differently. Roberts and Dunbar followed entire personal networks for 18 months across a major life transition and found that friendship quality declined, but that extra effort slowed the decline. For women, the strongest protection was more frequent contact, meaning talking together, while for men it was doing more activities together. This suggests that for many women the conversation itself is the maintenance work, so a friendship can be sustained by a call in a way that a vague plan to get together never sustains it. It also explains why a long silence can feel so costly between women who were once close, since the thing that held the friendship up was talking, and the talking stopped.

Life transitions rearrange the circle

Friendship networks do not shrink at random. A meta analysis of 277 studies with 177,635 participants found that the overall social network grows until young adulthood and then declines steadily, and that the friendship network shrinks throughout adulthood. Life events such as becoming a parent or starting a job produced network changes similar to those age patterns. In other words, a good part of what looks like personal failure to keep up is a pattern shared by large numbers of adults passing through the same milestones.

The milestones do different things to different relationships. Kalmijn, using twelve years of panel data, found that marriage leaves weaker ties alone but changes stronger ones, making friendships less important, while parenthood has negative effects on friendships and acquaintances that are partly offset by more contact with neighbors. The UC Berkeley Social Network Study found that younger adults experienced churn in their networks after a birth or a marriage or partnership, and that only the birth of a child was linked to lower confidence in receiving support. Churn means people leave the circle and others arrive, so a friendship that drops away after a wedding or a baby is often a casualty of turnover and not a verdict on the friendship.

Romance has its own measurable cost. In an internet sample of 540 respondents, Dunbar's team found that people in romantic relationships had smaller support networks, and that once the new partner was counted the cost came to nearly two members, spread equally between relatives and nonrelatives. Men and women were equally likely to lose close friends when a new relationship began. Dunbar's explanation is attention, since a new partner absorbs the hours that used to go elsewhere. That finding is useful because it separates the friend who drifts from the friend who has stopped caring. Most of the time the cause is a limited supply of hours, and the cause can be repaired.

Time is not shared out evenly

The hours that friendships need come out of the same day as everything else. The 2024 American Time Use Survey from the U.S. Bureau of Labor Statistics found that 87 percent of women and 74 percent of men spent some time on household activities on an average day. Men averaged 5.5 hours of leisure and sports activities a day while women averaged 4.7 hours. That gap of about 48 minutes a day adds up to roughly 290 hours over a year, which is more than the 200 hours Hall associated with a close friendship. These are averages across all adults and include leisure such as television, so they do not mean any one woman could redirect that time, but they do show that women begin the day with less free time to spend.

Friendship time has also been shrinking for everyone. An analysis of the American Time Use Survey from 2003 to 2020 found that social isolation rose and social engagement with family, friends, and others fell, and that work hours acted as a structural constraint on social engagement. A U.S. Surgeon General's graphic adapting those results shows time with friends falling by about 20 hours a month. The same analysis found that women spent more time with family while men spent more time with friends and in shared leisure. Taken together, those findings describe a woman whose hours are claimed first by work and household needs, and whose friendships receive what remains.

The broader picture is less bleak for women than for men, which is worth stating plainly. The Survey Center on American Life found that 41 percent of women reported six or more close friends in 1990 compared with 24 percent in 2021, and that 10 percent of women reported having no close friends, a smaller decline than men experienced. Women are keeping more friendships than men are, but they are keeping fewer than their mothers' generation did, and the people who are still in the circle are carrying more of the load.

Depth raises the stakes

Women's friendships tend to ask more of both people, which is part of their value. Hall's meta analysis found that women hold higher expectations of their same sex friends, although the difference is small. In Oswald, Clark, and Kelly's studies, female same sex friendships reported more supportiveness, openness, and interaction than male same sex friendships. A friendship built on frequent emotional exchange is rewarding when both people have the energy for it. It is also easy to let lapse when one person is depleted, because the friendship asks for more than a quick hello.

Talking about problems is where closeness and strain meet. Rose's 2002 study of 608 students found that girls spent more time than boys discussing and revisiting problems with friends, a pattern she called corumination, and that it was linked to both high quality friendships and more depression and anxiety symptoms. A later study of college students found that women reported more corumination with their closest friend, and that it predicted both higher friendship satisfaction and higher depressive symptoms. Rose has suggested the habit also takes up time that could go to more positive activities. The practical lesson is that a friendship can become close and heavy at once, and a tired woman may postpone a heavy conversation even when she wants the friend.

Effort also has to flow in both directions. Summaries of Oswald and colleagues' work report that reciprocal maintenance effort predicted how close a friendship became, and that when effort was not returned the friendship either rebalanced or deteriorated. In a study of 577 adults by Apostolou and Keramari, women were more likely than men to end a friendship over perceived selfishness. Women who initiate every plan may be quietly building resentment while the friend who never initiates feels no problem at all, and neither party sees the other's side.

The mind misreads the invitation

Even when a woman wants to reach out, her own mind works against her. Boothby and colleagues found that after conversations people systematically underestimate how much their partners liked them, and the gap persisted for months as college dorm mates built new relationships. Study co-author Gus Cooney reported that shy people were most prone to the gap. The cause appears to be that people substitute their own critical thoughts about themselves for what the other person thinks.

Reaching out is misjudged in a similar way. Across a series of preregistered experiments, Liu and colleagues found a robust tendency to underestimate how much other people appreciate being reached out to, partly because the person receiving the message focuses on the pleasant surprise of it. The underestimate held whether the gesture was a brief check in text or a small gift. It was larger when the contact was more surprising and when the two people were more socially distant. If two women each assume that reaching out would be an imposition, both wait, and the friendship decays on schedule without a single argument. That chain of events is my inference from these findings and not something the studies tested directly.

What a friend is worth

The case for the effort is not only sentimental. A 2010 meta analysis of 148 studies with 308,849 participants found a 50 percent greater likelihood of survival for people with stronger social relationships. The association was strongest for complex measures of social integration and weakest for simply living alone or with others. The review looked at social relationships broadly and not at friendship alone, so it does not prove that any one friend extends a life. It does show that the number and quality of connections carry weight that people tend to underrate.

How to keep a friendship going, one step at a time

The steps below turn the findings into routines. Where a step comes from a study, the study is named, and where it is a practical suggestion built on the findings, it is marked as one, because no trial has compared these exact routines. Nothing here requires more than a few hours a month. Start with step one and add the rest as each feels ordinary.

Step 1. Draw the circle on paper. Take a sheet of paper and write three headings. Under the first, list up to five women you would call in a crisis, which matches the inner layer Dunbar describes. Under the second, list up to fifteen women you enjoy and want to know better, and under the third, others you want to stay in touch with, up to about fifty. Write names only and do not judge the lists, since a first list with two names is a map of where your attention has been and not a score. Then put a star beside anyone you have not spoken to in three months, because the starred names are friendships already sliding down the layers.

Step 2. Give each circle a rhythm. Write a contact rhythm beside each heading. A practical starting point is some contact every week or two with the inner circle, about once a month with the middle circle, and a few times a year with the outer circle. Those intervals are a suggestion and not a research finding, so adjust them to the hours you have. The point of a rhythm is that decay is gradual and a steady small effort outperforms an occasional large one.

Step 3. Make contact by voice, not only by text. Roberts and Dunbar found that talking together protected friendships most for women, and Dunbar has noted that people generally find digital interactions less satisfying than face to face ones. A voice note, or a call of fifteen minutes during a walk or a commute, carries more of what a friendship runs on than a string of texts does. If a call feels like a big ask, use a text to open the door with a line such as "I have fifteen minutes on my drive Thursday, are you free for a quick catch up?" Once on the phone, three questions keep the conversation easy to start: what has taken up the most room in your head this week, what went well, and what would help right now.

Step 4. Send the first message without a reason. Liu's studies suggest the message will be appreciated more than you predict, so send it before the doubt wins. A message as plain as "I was thinking about you today and wanted to say hello, no need to reply right away" is enough. If a long gap has passed, one short acknowledgment is fine, such as "It has been too long and I miss you." Skip the paragraph of guilt, because a heavy apology makes the reply harder to write. Expect the answer to be warmer than your mind predicts, and treat the first reply as the start of a rhythm and not as a test.

Step 5. Book the next meeting before the current one ends. Hall's numbers show that closeness needs accumulated hours, and a standing plan protects those hours from the habit of waiting for someone to take the initiative. Before saying goodbye, pick a date, for example "first Thursday of the month, dinner at my place," and put it in both calendars. Choose a format that leaves room to talk, since Hall found that hours spent working together counted for less. A recurring walk, a shared meal, or an errand run together can all work, and the ritual matters more than the venue.

Step 6. Use the four maintenance behaviors as a checklist. Oswald and colleagues identified positivity, supportiveness, openness, and interaction as the core friendship maintenance behaviors, and found that both friends' behaviors predicted satisfaction and commitment for the pair. In practice, positivity means celebrating her news within a day, supportiveness means asking what kind of help she wants before offering solutions, openness means sharing one honest thing about your own week including something unresolved, and interaction means the contact rhythm from step two. Pick one behavior per conversation and aim to practice all four over a month. Keep it light, since the goal is consistency and not a performance.

Step 7. Keep problem talk from taking over. Corumination research suggests that long, repetitive problem sessions can bring friends closer and also leave both women more anxious. Give a problem its full turn, then ask a question such as "Do you want to think through options or do you mainly need me to listen tonight?" That question gives her control and keeps you from guessing. When the problem has had its time, move deliberately toward a plan, a memory, or something funny. This step is my practical application of Rose's findings and not an intervention that has been tested.

Step 8. Name lopsided effort early and gently. When you are the one always reaching out, start by assuming the cause is overload, because the transition research shows how many ordinary life events consume a friend's attention. Then say it kindly and specifically, for example "I have noticed I have been the one reaching out lately, and I miss you. Could we set a time this month?" Ask once and propose a date. If effort stays one sided for months, it is reasonable to move her to a lighter circle on your list, which matches how Dunbar describes friendships changing layers, and to keep the door open without resentment.

What the research cannot tell us

These findings have limits worth knowing. Hall's hour estimates came from adults who had recently moved and from college freshmen, so they are a useful guide and not a law. The sex differences in friendship expectations are small, which means the individual woman matters more than the group average. Much of the corumination research began with children and adolescents, the Survey Center data is self reported, and most of the studies cited here drew on samples from Western countries. The steps above are reasoned applications of the findings, and they are offered as things to try and adjust.

Starting with one name

A friendship that has gone quiet has usually not closed. The research describes a slow, ordinary drift driven by attention and hours, and it points to a remedy that is just as ordinary: a call this week, a date on the calendar, a message sent without a reason. Close friendships are kept in small repeated acts that no one else sees. Look at the starred names on your list, choose one, and send the first message tonight.