Workplace Loneliness Statistics 2026: The Belonging Gap, in Verified Numbers
There is no single "workplace loneliness rate", and that is the first thing anyone citing these statistics needs to understand. Credible, well-run studies published between 2023 and 2026 put the figure at 19%, at 40%, at 52% and at 85%. None of them is wrong. They are asking materially different questions of different populations over different time windows, and the answers are not interchangeable.
This page assembles the verified 2026 dataset on loneliness, exclusion and belonging at work. It states what each figure actually measures, says plainly where the evidence is strong and where it is thin, and flags the widely-circulated numbers we could not trace to a primary source. Every figure is attributed in the Methodology and sources section.
1. How many employees are actually lonely? Why the answer ranges from 19% to 85%
Workplace loneliness is measured in at least three incompatible ways, and the choice of method moves the headline number by more than sixty percentage points.
Yesterday-experience measures ask whether you felt lonely for a lot of the day yesterday. This is Gallup's method, and it produces the lowest figures — 19% of US employees in the 2026 edition of State of the Global Workplace. It is a strict test: it captures acute, present-tense loneliness on a single ordinary day, not a general disposition.
Self-assessment measures ask whether you consider yourself lonely, usually without a time bound. Cigna and the Evernorth Research Institute, surveying more than 7,500 US adults, found 57% of Americans overall and 52% of workers report loneliness on this basis. Perceptyx, asking employees whether they feel "very" or "somewhat" lonely at work specifically, found more than four in ten.
Lifetime-incidence measures ask whether you have ever experienced something. EY's Belonging Barometer reports that 75% of workers have felt excluded at work, rising to around 85% reporting moments of exclusion in the most recent wave. These are the largest and most quoted numbers, and the most frequently misused — "85% of employees are lonely" is not what that data says.
The same phenomenon, measured five ways
Sources: Gallup State of the Global Workplace 2026 and 2025; Perceptyx (2023, n = 2,000+); Cigna Group/Evernorth Research Institute, Loneliness in America 2025 (n = 7,500+); EY Belonging Barometer. Bars are scaled to a common axis for comparison only — these are not like-for-like measures.
2. Loneliness by work arrangement
Gallup's 2025 edition broke daily loneliness out by work arrangement, and the gradient is consistent with what you would expect — though the spread is narrower than the remote-work debate usually implies. Seven percentage points separate fully remote employees from on-site remote-capable ones. Remote work is associated with more loneliness; it is not the sole cause of it, and one in five on-site employees reports the same thing while surrounded by colleagues.
Daily loneliness by work arrangement
Share of employees who felt lonely a lot of the previous day. Source: Gallup, State of the Global Workplace 2025. Remote-capable employees only.
Loneliness does not travel alone. Among fully remote workers in the same dataset, 30% reported sadness and 25% reported anger the previous day — the highest of any work arrangement on all three measures. The uncomfortable finding for anyone using engagement as a proxy for wellbeing is that fully remote employees are simultaneously among the most engaged and the least thriving. Engagement and wellbeing are different constructs and remote work pulls them in opposite directions. Our hybrid and remote work statistics post covers that divergence in more depth.
Negative daily emotions among fully remote employees
Share reporting each emotion for a lot of the previous day. Source: Gallup, State of the Global Workplace 2025.
3. The loneliest people in the building are the ones running it
This is the most counterintuitive finding in the dataset, and it survives across sources. Perceptyx's research found senior leaders roughly twice as likely as lower-level managers and individual contributors to describe themselves as very lonely at work. Gallup's more recent work points the same direction, with senior leadership scoring meaningfully higher on loneliness than individual contributors.
Seniority buys you more meetings and fewer peers. The org chart that concentrates decision rights also concentrates isolation.
The mechanism is structural rather than personal. As people move up, the number of colleagues they can speak to as equals shrinks, the proportion of their interactions that are evaluative rather than social rises, and candour becomes professionally expensive. Perceptyx also found men roughly twice as likely as women to report being very lonely — a gap consistent with the broader literature on male friendship networks contracting through mid-career.
The practical consequence is that most recognition and connection programmes are aimed at exactly the population that needs them least. Peer recognition schemes are typically designed for individual contributors; senior leaders are cast as the givers of recognition and almost never as recipients. If your programme has a "who recognises the leadership team?" answer of nobody, that is a measurable gap, not a philosophical one.
4. Loneliness by generation
Both Cigna/Evernorth and Perceptyx find Gen Z and Millennial employees lonelier than Gen X and Baby Boomers — despite being the most technologically connected cohorts ever to enter the workforce. In Gallup's regional cuts, loneliness among younger workers in the US and Canada climbs to around 26%, well above the 19% national employee figure.
The pattern matters for programme design because early-career employees are also the group with the shortest tenure, the weakest internal networks and the highest voluntary turnover. Loneliness in that cohort converts into attrition faster than in any other. See our Gen Z workplace statistics and employee turnover statistics for the adjacent data.
5. What loneliness costs employers
The most-cited employer-side figure comes from the 2023 US Surgeon General's advisory Our Epidemic of Loneliness and Isolation, which put the annual cost to American employers of stress-related absenteeism attributable to loneliness at approximately $154 billion. The underlying mechanism is straightforward: lonely workers in the cited research missed more than five additional working days per year than workers who were not lonely.
Perceptyx quantified the performance side. Compared with their connected colleagues, lonely employees were roughly 1.5× more likely to be disengaged and 4.5× more likely to struggle with productivity, and between four and six times more likely to report difficulty sleeping, unhealthy coping behaviours and being bothered by physical pain. Cigna found less lonely employees more likely to say they work hard than lonely ones (74% versus 63%).
| Effect | Finding | Source |
|---|---|---|
| Absenteeism cost | ~$154bn per year to US employers (stress-related absenteeism) | US Surgeon General, 2023 |
| Days lost | 5+ additional missed working days per year per lonely worker | Research cited in the Surgeon General's advisory |
| Disengagement | Lonely employees ~1.5× more likely to be disengaged | Perceptyx, 2023 |
| Productivity | Lonely employees ~4.5× more likely to struggle with productivity | Perceptyx, 2023 |
| Discretionary effort | 74% of less lonely employees work hard, versus 63% of lonely employees | Cigna/Evernorth, 2025 |
| Retention | Lonely workers more likely to seek opportunities elsewhere | Cigna/Evernorth, 2025 |
For context on the size of the disengagement problem these numbers sit inside: Gallup puts global employee engagement at 20% in 2025 and the cost of low engagement at more than $10 trillion in lost productivity, or roughly 9% of global GDP. Loneliness is one input to that, not the whole of it.
6. Belonging: the same problem, measured from the other end
Loneliness data tells you what is wrong. Belonging data tells you what it is worth fixing. BetterUp's research — 1,789 full-time employees across industries, supplemented with live participant experiments — produced the effect sizes most often used to build the business case.
| Outcome | Strong sense of belonging vs. weak |
|---|---|
| Job performance | +56% |
| Turnover risk | −50% |
| Sick days taken | −75% |
| Likelihood of recommending the employer | +167% |
| Modelled productivity value, 10,000-person company | >$52m per year |
BetterUp also measured the downside with unusual precision: a single incident of micro-exclusion — being talked over, left off an invitation, having a contribution reattributed — produced a 25% decline in that individual's performance on a subsequent team task. That is the smallest unit of harm in this entire dataset, and it is the one most under management control.
EY's Belonging Barometer supplies the prevalence side. Around 41% of workers globally say the workplace is where they feel their greatest sense of belonging, second only to home — workplaces are genuinely important sources of connection, not merely places where connection is lost. But 75% report having felt excluded at work, and 56% say they cannot or are reluctant to share dimensions of their identity at work for fear it will hold them back.
7. Recognition is the intervention with published effect sizes
Most proposed remedies for workplace loneliness — social committees, offsites, mandatory in-office days — have no published effect size against loneliness at all. Recognition does, from a joint Workhuman and Gallup research programme that tracked the career paths of more than 3,400 employees between 2022 and 2024.
| Outcome | Finding |
|---|---|
| Daily loneliness | 66% less likely, where recognition fulfils four or more quality pillars |
| Sense of belonging | Up to 10× as likely to strongly agree they belong at the organisation |
| Meaningful connection at work | Up to 7× as likely to strongly agree they have a best friend or meaningful connection |
| Sense of purpose | 4.4× more likely to strongly agree their job gives them purpose in life |
| Voluntary turnover | Employees receiving high-quality recognition in 2022 were 45% less likely to have left by 2024 |
The qualifier matters more than the headline. The effect is attached to recognition that meets four or more of five quality pillars — fulfilling, authentic, personalised, equitable, and embedded in the culture. Recognition that fails those tests is not a weaker version of the same intervention; the research does not support extending these effect sizes to a generic quarterly all-hands shout-out or an automated work-anniversary email.
"Equitable" is the pillar most programmes fail, and it is the one that bears directly on the loneliness data above. If recognition concentrates on the visible, the co-located and the already well-networked, it will widen the belonging gap for exactly the remote, early-career and peripheral employees who report the most loneliness. Our guides to peer-to-peer recognition and frontline employee recognition cover the distribution problem in practice, and employee recognition statistics 2026 has the broader evidence base.
8. Workplace loneliness and belonging cheat sheet (2026)
| Metric | Figure | What it measures | Source |
|---|---|---|---|
| US employee daily loneliness | 19% | Felt lonely a lot of the previous day | Gallup, 2026 |
| Fully remote daily loneliness | 27% | Felt lonely a lot of the previous day | Gallup, 2025 |
| Hybrid daily loneliness | 23% | Felt lonely a lot of the previous day | Gallup, 2025 |
| On-site daily loneliness | 20% | Felt lonely a lot of the previous day (remote-capable) | Gallup, 2025 |
| Lonely at work | 40%+ | "Very" or "somewhat" lonely at work | Perceptyx, 2023 |
| US workers reporting loneliness | 52% | Self-identified loneliness, no time bound | Cigna/Evernorth, 2025 |
| US adults reporting loneliness | 57% | Self-identified loneliness, general population | Cigna/Evernorth, 2025 |
| Have felt excluded at work | 75% | Lifetime incidence | EY Belonging Barometer |
| Workplace as greatest source of belonging | 41% | Ranked second after home | EY Belonging Barometer 3.0 |
| Reluctant to share identity at work | 56% | Fear it would hold them back | EY Belonging Barometer |
| Global employee engagement | 20% | Gallup engagement definition, 2025 data | Gallup, 2026 |
| Global daily stress | 40% (US: 50%) | Felt a lot of stress the previous day | Gallup, 2026 |
| Global thriving | 34% | Life evaluation "thriving" | Gallup, 2026 |
9. Loneliness statistics to handle with care
10. What to actually do with this data
Measure loneliness directly, once
Most engagement surveys have no loneliness item at all. Add the strict Gallup-style question — did you feel lonely a lot of the day yesterday — to your next pulse, and cut it by work arrangement, tenure and level. You cannot manage a 19–27% problem you have never measured.
Audit recognition for equity, not volume
Pull the distribution of recognition events by team, location and work arrangement. If remote and frontline employees receive proportionally less, your programme is currently widening the belonging gap that the loneliness data describes.
Include the leadership layer
Senior leaders report roughly double the rate of acute loneliness and are structurally excluded from peer recognition. Name who recognises them. An unanswered question here is a real gap, not a technicality.
Target micro-exclusion, not just events
A single micro-exclusion incident cost 25% of subsequent task performance in BetterUp's experiments. Meeting norms on airtime, credit attribution and invitation lists are cheaper than any offsite and act on the mechanism directly.
Fix the early-career cohort first
Gen Z and Millennial employees are lonelier on every measure and have the shortest tenure and weakest internal networks. That combination converts loneliness into attrition faster than in any other group.
Do not sell RTO as the cure
On-site employees still report 20% daily loneliness and fully remote employees are the most engaged group. Mandating attendance addresses co-location, not connection, and the data will not back the business case afterwards.
Methodology and sources
Every figure on this page is attributed to a named study, with sample size and fieldwork period stated where published. Where sources disagree, both figures are shown with the question each asked. Figures we could not trace to a primary source are listed in section 9 rather than repeated as fact.
Key takeaways
- There is no single workplace loneliness rate. Cite the question with the number: 19% is daily prevalence, 52% is self-identification, 75% is lifetime exclusion.
- Remote work is associated with more loneliness, but the spread is seven points and on-site employees still report 20%. Co-location is not connection.
- Senior leaders and early-career employees are the two loneliest groups, and most connection programmes are designed for neither.
- Belonging carries large published effect sizes on performance, turnover and absence — and micro-exclusion carries a measurable cost at the level of a single incident.
- Recognition is the only common intervention here with a published effect size specifically against daily loneliness, and only when it is personalised, authentic, equitable and embedded rather than generic.
Make recognition reach the people the data says are loneliest
Equitable recognition is the pillar most programmes fail — and remote, frontline and early-career employees are the ones it fails first. Rewordin delivers gift-card rewards in 150+ countries in local currencies, so a distributed team gets the same recognition experience as the people in the head office, with the delivery and redemption reporting to prove it landed.
About the authors
Maciej is the founder and CEO of Rewordin, a global employee rewards and recognition platform delivering gift cards in 150+ countries. He works directly with HR, People Ops and finance teams on reward programme design, HR systems integration and bulk gift card procurement, and writes about the research behind effective recognition. Based in Wrocław, Poland. Connect on LinkedIn →
Natalia is the CFO of Rewordin and co-reviewer of every cost and market claim published on the platform — including the absenteeism cost estimate and the modelled belonging productivity figures set out in this report, both of which are presented here with their scope and limitations stated. Connect on LinkedIn →
What percentage of employees are lonely at work in 2026?
It depends entirely on the question asked. Gallup's State of the Global Workplace 2026 found 19% of US employees felt lonely a lot of the previous day — a strict, present-tense measure. Cigna and the Evernorth Research Institute, asking US workers whether they consider themselves lonely, found 52%. Perceptyx, asking whether employees feel very or somewhat lonely at work, found more than four in ten. EY reports 75% have felt excluded at work at some point, which is a lifetime measure rather than a current rate. All four are credible; they are not interchangeable.
Are remote workers lonelier than office workers?
Yes, but by less than most people assume. Gallup's 2025 edition put daily loneliness at 27% for exclusively remote employees, 23% for hybrid and 20% for on-site remote-capable employees — a seven-point spread. Fully remote employees also report the highest rates of sadness (30%) and anger (25%), while simultaneously being among the most engaged and least thriving. One in five on-site employees reports the same daily loneliness, so proximity alone does not resolve it.
Who is loneliest at work?
Two groups stand out. Senior leaders are roughly twice as likely as lower-level managers and individual contributors to describe themselves as very lonely at work, according to Perceptyx — a structural effect of having fewer peers and more evaluative interactions as seniority rises. Gen Z and Millennial employees are also lonelier than Gen X and Baby Boomers across both the Cigna and Perceptyx datasets, with loneliness among younger US and Canadian workers reaching around 26% on Gallup's daily measure.
What does workplace loneliness cost employers?
The 2023 US Surgeon General's advisory estimated approximately $154 billion annually in stress-related absenteeism attributable to loneliness among American employers, with lonely workers missing more than five additional working days per year. Perceptyx found lonely employees roughly 1.5 times more likely to be disengaged and 4.5 times more likely to struggle with productivity. Note that the $154bn figure is a national estimate for one specific cost channel — dividing it by headcount to produce a per-company number is not defensible.
Does employee recognition actually reduce loneliness?
It is the intervention with the clearest published effect size. Workhuman and Gallup, tracking more than 3,400 employees from 2022 to 2024, found that employees whose recognition fulfils four or more of five quality pillars are 66% less likely to experience daily loneliness, up to 10 times as likely to strongly agree they belong, and up to 7 times as likely to report a meaningful connection at work. The effect is conditional on quality: the five pillars are that recognition be fulfilling, authentic, personalised, equitable and embedded in the culture. Generic or automated recognition is not covered by these findings.
What is the difference between loneliness and belonging at work?
Loneliness is how an employee feels about the quality of their connections; belonging is whether they judge themselves to be an accepted member of the group. They correlate strongly but respond to different interventions. Social events can reduce loneliness for well-connected employees while increasing it for those already on the periphery, whereas inclusion practices — equitable airtime, credit attribution, consistent recognition — act on belonging directly. BetterUp's data shows why belonging is worth measuring separately: employees with a strong sense of belonging showed 56% higher job performance and 50% lower turnover risk than those with a weak one.
Will a return-to-office mandate fix workplace loneliness?
The data does not support that case. On-site remote-capable employees still report 20% daily loneliness in Gallup's figures, only seven points below fully remote employees, and fully remote employees are among the most engaged groups measured. Mandating attendance addresses co-location rather than the quality of interaction, which is what the loneliness and exclusion research consistently identifies as the mechanism. Micro-exclusion — being talked over, left off invitations, having contributions reattributed — happens in offices as readily as on video calls, and BetterUp measured a 25% performance decline from a single incident of it.