.png)
The Chief of Staff Is Lonely by Design, and the Usual Advice Makes It Worse
There is a moment, usually about four months into a Chief of Staff role, when somebody senior closes the door, tells you something you are not allowed to know yet, and leaves. You now know a thing. Nobody else on your floor knows it. You will sit in three meetings this week where it is relevant and you will say nothing, and you will do that well, because that is the job.
Then you will go home and think about it.
The old version of this post, written in 2023, said that being a Chief of Staff is lonely and that you should build a support network, find a mentor, and connect with colleagues on a personal level. I stand by the first half. The second half is what everybody says, and when I went and read the evidence properly, it turned out to be the weakest advice available, and in one important way it is backwards.
What you will get out of this
The three separate lonelinesses the job manufactures, and why they are structural rather than personal
What the research on secrecy says about why confidentiality is tiring, which is not the reason you think
The mortality number everybody quotes about loneliness, the smaller one they leave out, and the zero underneath both
Why "spend more time with people" is measurably the wrong lever, and what moves the needle instead
The one relationship worth building, what to ask of the person, and what never to tell them
A one-page brief to hand to the person you choose
The three lonelinesses the job builds
Most writing about loneliness at work treats it as one thing that happens to sensitive people. The literature does not. Since the 1970s it has distinguished emotional loneliness, which is the absence of a close attachment, from social loneliness, which is the absence of a peer group. You can have one without the other. A person with a loving partner and no friends at work is socially lonely. A person surrounded by a friendly team with nobody they can actually talk to is emotionally lonely.
The Chief of Staff role is unusual in that it manufactures both at once, through three separate mechanisms. None of them are about you.
The confidant with no confidant. You are told things. Restructures before they are announced. Which of two people is leaving. What the board actually said. A large study of who gets confided in found that compassion and assertiveness predict being told secrets, and politeness and enthusiasm predict being told fewer. That is a description of the person hired into this role. You are selected for exactly the traits that make people close the door, and then you are contractually forbidden from ever doing the same.
The peer with no peers. There is typically one of you. The people at your level do not do your job, and the person whose job you are closest to is the person you work for, which is not a peer relationship however friendly it is. Interviews with senior executives found that loneliness at the top runs on three things: increased social distance, a lack of social support, and exhaustion from the role. A Chief of Staff inherits the first two from the principal without inheriting the title.
The person mistaken for the role. Those same interviews identified a conflict they called Role versus Person. People respond to what you represent, which is access, which is the principal's ear, which is a channel. It is hard to have an ordinary conversation with someone who is calculating whether it will get back to your boss. Over time the ordinary conversations stop being offered.
You can build a support network and none of these change. They are properties of where the role sits, not of how sociable you are.
Why confidentiality is tiring, and it is not the reason you think
For a long time the assumption was that keeping secrets is hard because concealing them is hard. You have to watch what you say, monitor the conversation, steer away from the subject. Effortful, and therefore draining.
That turns out to be mostly wrong. A series of studies analysing more than 13,000 secrets found that people catch themselves spontaneously thinking about their secrets far more often than they meet a situation that actually requires hiding them. And it is the thinking, not the hiding, that predicts lower wellbeing. The frequency of concealment did not.
A follow-up set of seven experiments pinned down the mechanism. Thinking about a secret, compared with thinking about personal information that simply has not come up, increases fatigue by evoking feelings of isolation and a conflict with the person's need to belong. The fatigue showed up in task persistence and performance.
Read that again with the job in mind. The cost of confidentiality is not paid in the meeting where you keep your mouth shut. You are good at that. It is paid at nine in the evening, alone, when the thing you know comes back into your head and brings a small feeling of separateness with it. Every time.
There is a widely quoted finding that people carrying a heavy secret judge hills to be steeper. I would love to use it. It did not replicate, and a careful attempt in 2015 found no effect at all, so it stays out.
The number everybody quotes, and the two underneath it
If you have read anything about loneliness in the last decade, you have seen the mortality figure. A meta-analysis of studies from 1980 to 2014 found loneliness associated with a 26% increase in the likelihood of death, roughly comparable to well-established risk factors. It is a real finding and it is everywhere.
Here is what sits underneath it. A larger and more recent review of 90 cohort studies covering more than two million people put the loneliness figure at 14%, and social isolation, which is the objective lack of contact rather than the feeling, at 32%. And the largest single cohort I could find, nearly half a million people in one national biobank, found that once you account for everything that travels with loneliness, income, health behaviours, existing illness, mood, the independent effect of loneliness on mortality was zero. Not small. Zero. An earlier study of 6,500 older adults found the same pattern: isolation stayed associated with death, loneliness did not.
I am not telling you loneliness is harmless. I am telling you the headline figure is the least useful thing about it for a working professional, and it is quoted because it is frightening rather than because it is relevant.
The relevant evidence is smaller and duller. A 2023 meta-analysis of 49 workplace studies found loneliness at work associated with lower job performance, lower job satisfaction, a worse relationship with one's manager, and higher burnout. Correlations run from zero, no relationship, to one, perfect lockstep, and these landed at about a third, which for human behaviour is a real relationship and not a rounding error. The studies were mostly snapshots, so nobody can say which direction it runs. It probably runs both.
That is the finding to carry. Not that loneliness will shorten your life, but that it is quietly taxing the thing you were hired for.
The uncomfortable claim underneath all of this
Every piece of work like this needs one claim in the middle where the evidence contradicts what everybody knows. Here it is.
More social contact is the weakest lever available, and for a lonely person it can make things worse.
Three findings. First, the intervention literature. The interventions that have been tested for loneliness fall into four types: teaching social skills, adding social support, creating more opportunities for contact, and changing how a person reads their social world. The first meta-analysis to compare them found that, in properly randomised trials, the last of those was the one that worked. An umbrella review in 2020 put numbers on it: adding social support moved loneliness by a small amount that barely cleared zero, while training in how people interpret social situations moved it about four times as far. A review of 280 studies published this year reached the same ordering.
Second, the experience-sampling work. Three datasets, more than three thousand people, pinged through the day and asked how they felt and who they were with. The negative link between loneliness and wellbeing was stronger when people were with others than when they were alone. At high loneliness, being in company was associated with the same wellbeing as being alone, or lower. The authors' words were that simply spending time with others "is not associated with a reduced burden of loneliness and may even backfire".
Third, the secrecy work again. Confiding a secret helps, but not the way you would expect. It does not reduce how often the secret has to be concealed. It helps because the person feels they have support and are able to cope, and that feeling predicts less time spent alone with the thought. The relief comes from the sense of being able to carry the thing, not from putting it down.
Put those together and the standard advice collapses. Building a bigger network adds contact, and contact is the lever with the smallest effect. Connecting with colleagues on a personal level adds people who, by the third mechanism above, are not entirely sure whether they are talking to you or to your boss. And "sharing more" is, for this role, both forbidden and beside the point.
What actually helps
The evidence points somewhere quite specific, and it is not where the old post pointed.
One confidant for the weight, not the content. The research says the harm of a secret is the isolation of thinking about it alone, and the relief of confiding comes from feeling able to cope, not from the other person knowing what it is. Which means you can get almost all of the benefit without breaking a single confidence. You need one person to whom you can say "I am carrying something heavy this week and I cannot tell you what it is," and who will not ask. That sentence is enough. It converts a secret you are alone with into a weight somebody knows you are holding.
Choose them by the same traits that got you the job. Compassion and assertiveness predict being confided in. Politeness and enthusiasm predict the opposite. You want someone who will take it seriously and will not fill the silence with reassurance. They do not need to be senior and they do not need to understand your industry. They need to be able to sit with the fact that you are not going to explain.
Remember why you are keeping it. The secrecy research found that secrets kept for a professional or goal-oriented reason are experienced very differently from secrets carrying shame, and that reminding people of the reason a secret is kept improves their sense of being able to cope with it. Professional confidentiality carries no shame. It is the least harmful kind of secret there is, as long as you do not let it feel like the other kind. When it comes back at nine in the evening, the useful thought is not "I cannot tell anyone." It is "I am holding this because somebody trusted me to, and I am good at it."
Read the room differently. The interventions that work are the ones that change how a lonely person interprets social signals, because loneliness makes people hypervigilant for rejection and quick to read it where it is not. The Role versus Person conflict makes this worse: you are actually being treated as a channel some of the time, which makes it very easy to assume you are being treated that way all of the time. The correction is boring and effective. When a conversation feels careful, ask yourself whether the person is managing you or managing what you represent. Most of the time it is the second, and the second is not about you.
Stop journalling about it. The old post recommended it. The evidence on writing about your feelings, pooled across 146 randomised studies, found an average effect a whisker above zero, and a separate review of 30 trials found no effect at all. It is close to harmless and close to useless. The difference between writing it down and saying it to a confidant is that the second one involves another person knowing you are carrying it, and that is the whole mechanism.
Download: The Confidant Brief
One page, and it is not for you. It is for the person you choose.
Asking somebody to be the person you tell nothing to is an odd request, and most people will try to help by asking questions. The brief does the explaining so you do not have to.
What is inside | Why it is there |
What you are asking for, in three sentences | Because "can I talk to you sometimes" gets a yes and then a lot of questions |
The one sentence you will say, and the one reply that helps | Because confiding works through feeling able to cope, and the reply that does that is shorter than people think |
What they will never be told, and why that is not a lack of trust | Because the harm is in carrying it alone, not in them not knowing |
A short reading list from this piece | Because some people will only agree once they understand it is not therapy |
Free, no email wall. Print it, hand it over, and let the page have the awkward conversation.
If you are about to take the role
The 2023 version of this post was written for the person considering the job, and I want to keep that.
Nothing above is a reason not to take it. The role is genuinely rewarding and the loneliness is genuinely built in, and both of those are true at once. What I would say now, that I did not say then, is that the loneliness is not a sign the job is going badly or that you are doing it wrong. It is a sign the job is working. You are being told things because you can be trusted with them, and you are being kept at a distance because you are close to power. Both are compliments in an inconvenient shape.
What I would do before starting is find the confidant first. Not a mentor, not a network, not a coach, though have those too if you want them. One person, briefed properly, who knows the sentence "I cannot tell you what it is" is coming and has already agreed not to mind.
Frequently asked questions
Is this different from ordinary workplace loneliness?
In kind, yes. Most workplace loneliness comes from a lack of connection. This comes from being over-connected in one direction: everyone confides in you, and the structure forbids the return journey. The research on keeping other people's secrets found that being confided in brings closeness and burden together, and the closer you are to the person confiding, the more of both. A Chief of Staff is as close as it gets.
Can I use my partner as the confidant?
You can, and many people do. Two cautions from the evidence. The closer the confidant is to your world, the more they will think about what you are carrying, which is a burden for them too. And a partner is more likely to ask, because they care. If they can hear "I cannot say" and leave it there, they are ideal. If they cannot, it is kinder to both of you to choose somebody else.
What about a coach or a therapist?
A coach is a perfectly good confidant for the weight, and a professional one is used to not being told things. I am deliberately not giving advice about therapy in a blog post about a job. If the loneliness is bigger than the job, that is a different conversation with a different person.
Does my principal know this is happening?
Usually not. The interviews with executives found they were often lonely themselves and rarely aware that the people closest to them were too. You do not need to raise it with them, and I would not. The confidant is the fix, not the boss.
Will this get better as I get more senior?
The evidence suggests not on its own. Loneliness clusters and spreads through networks, up to three degrees of separation, and it concentrates at the edges. Seniority tends to move you further from the centre of the social network rather than closer. The fix is the same at every level.
Is it really zero? The mortality thing?
In that cohort, once every other factor was accounted for, yes. In the bigger pooled review it was 14%. In the famous one it was 26%. All three are honest numbers from honest studies. The difference is what they controlled for, and the lesson is that loneliness travels with other things that do the damage. Which is a reason to take it seriously and not a reason to be frightened by it.
Where to start
Do not build a network this week. Write down the name of one person who could hear "I am carrying something and I cannot tell you what" without asking a follow-up question. If nobody comes to mind, that is the finding, and it is a more useful one than any number in this piece.
Then give them the brief.
Most of what this role asks of you, holding what you are told, staying steady when the room is careful with you, protecting your own capacity while you protect everybody else's, sits inside what the Connected Leader Assessment calls Sustainable Performance. It is free, it takes about twelve minutes, and it will tell you whether the thing you are carrying is being carried well or is quietly costing you.
And if you want the wider territory, how to hold the room, say no, and have the conversation you have been putting off, that is what our Skill Sprints are for. One topic, built properly, an hour or two, and you come out having made something rather than having been talked at.
Meg ✌️
Free tools
The weight check, the interactive tool in this post, free
The Confidant Brief, the download in this post, free
The Connected Leader Assessment, twelve minutes, free, with a read on Sustainable Performance, the dimension this post sits in
P.S. Want to Go Deeper on Secrets and Solitude?
If this resonated and you want to go further, here is the evidence the piece is built on. These are the papers that shaped my thinking on why this role is lonely by design and what the research says actually helps, and they are well worth your time.
Every DOI below has been checked. Where a popular claim did not survive that check, I have said so in the piece rather than quietly leaving it out.
📃 Slepian, Chun and Mason (2017) The experience of secrecy. More than 13,000 secrets, and the finding that thinking about a secret, not concealing it, is what predicts lower wellbeing. https://doi.org/10.1037/pspa0000085
📃 Slepian, Halevy and Galinsky (2018) The solitude of secrecy. Seven experiments showing that thinking about a secret evokes isolation and fatigue, with consequences for persistence and performance. https://doi.org/10.1177/0146167218810770
📃 Slepian and Moulton-Tetlock (2018) Confiding secrets and wellbeing. Confiding helps through feeling able to cope, not through reduced concealment. https://doi.org/10.1177/1948550618765069
📃 Slepian and Greenaway (2018) The benefits and burdens of keeping others' secrets. Being confided in brings closeness and burden together. https://doi.org/10.1016/j.jesp.2018.02.005
📃 Slepian and Kirby (2018) To whom do we confide our secrets? Compassion and assertiveness predict being confided in; politeness and enthusiasm predict the opposite. https://doi.org/10.1177/0146167218756032
📃 Slepian, Kirby and Kalokerinos (2020) Shame, guilt and secrets on the mind. Shame drives mind-wandering to a secret; guilt reduces it. https://doi.org/10.1037/emo0000542
📃 Pecher, van Mierlo, Cañal-Bruland and Zeelenberg (2015) The failed replication of the "secrets make hills look steeper" finding. https://doi.org/10.1037/xge0000090
📃 Bryan, Andrews, Thompson, Qualter, Matthews and Arseneault (2023) The workplace loneliness meta-analysis, 49 studies: lower performance, lower satisfaction, higher burnout. https://doi.org/10.1093/occmed/kqad138
📃 Zumaeta (2018) Lonely at the top. Interviews with senior executives, and the Role versus Person conflict. https://doi.org/10.1177/1548051818774548
📃 Silard and Wright (2022) Distinctly lonely. How the pathways to loneliness differ between managers and their employees. https://doi.org/10.1108/mrr-05-2021-0379
📃 Russell, Cutrona, Rose and Yurko (1984) Emotional and social loneliness are distinct, and you can have both. https://doi.org/10.1037/0022-3514.46.6.1313
📃 Holt-Lunstad, Smith, Baker, Harris and Stephenson (2015) The meta-analysis behind the 26% figure. https://doi.org/10.1177/1745691614568352
📃 Wang and colleagues (2023) 90 cohort studies, 2.2 million people, and the 14% figure. https://doi.org/10.1038/s41562-023-01617-6
📃 Elovainio and colleagues (2017) The biobank cohort in which the independent effect of loneliness on mortality went to zero after full adjustment. https://doi.org/10.1016/s2468-2667(17)30075-0
📃 Steptoe, Shankar, Demakakos and Wardle (2013) Isolation predicted mortality; loneliness did not, once health and demographics were included. https://doi.org/10.1073/pnas.1219686110
📃 Cacioppo, Fowler and Christakis (2009) Alone in the crowd. Loneliness clusters, spreads to three degrees, and sits at the edges of networks. https://doi.org/10.1037/a0016076
📃 Cacioppo and Hawkley (2009) Why loneliness changes how social information is read, and the mechanism behind the interventions that work. https://doi.org/10.1016/j.tics.2009.06.005
📃 Stavrova and Ren (2023) Alone in a crowd. Three experience-sampling datasets showing that time with others does not reduce the burden of loneliness and may backfire. https://doi.org/10.1007/s10902-023-00661-3
📃 Masi, Chen, Hawkley and Cacioppo (2011) The first meta-analysis of loneliness interventions. In randomised trials, changing social cognition worked best. https://doi.org/10.1177/1088868310377394
📃 Veronese and colleagues (2020) The umbrella review with the numbers: social support barely clears zero, social cognitive training moves loneliness about four times as far. https://doi.org/10.1111/hsc.13248
📃 Lasgaard and colleagues (2025) 280 studies. Psychological interventions most effective; support and network interventions small to moderate. https://doi.org/10.1037/amp0001578
📃 Frattaroli (2006) 146 randomised studies of writing about your feelings, average effect a whisker above zero. https://doi.org/10.1037/0033-2909.132.6.823
P.P.S. Why Do I Even Have the Nerve to Write This?
Fair question.
I am someone who has spent years working alongside remarkable executives, navigating chaos, translating vision into structure and figuring things out, sometimes beautifully and sometimes the very hard way. I have had the privilege of being the steady right hand to leaders who move fast, take risks and expect a lot. That environment has shaped me more than any classroom ever could.
I also happen to have an MBA, and I am studying psychology because people fascinate me. How we work. Why we disconnect. What actually holds us together when the pressure rises.
But really, none of that is the point.
I am here because I have lived the erosion and rebuilt from it. I have seen the cost of carrying too much and the relief that arrives when you finally stop disappearing inside your own competence.
So I share what I have learned in case it helps someone else. Take what is useful. Leave the rest. And remember this one truth that professionals like you often forget:
You are probably doing far better than you give yourself credit for.
Comments