Four well-being stories: how HR, leaders, managers, and employees are doing, and what's stressing them

The prior YoY Trends 2026 study reported that the platform-wide WHO-5 score rose six percent over 24 months while engagement signals collapsed. The follow-up question: whose well-being moved, and are happiness and stress travelling in the same direction underneath that headline?

To answer it we split 6,069 active users across 117 companies into four mutually exclusive role groups, then ran each of the three signals over the same 24-month window. The groups are HR (people in the HR function), Leader (top of an org chart, typically a CEO, founder, or managing director, with at least one direct report), Manager (has at least one direct report and is not a Leader or in HR), and Employee (everyone else). Priority order is HR, then Leader, then Manager, then Employee, so an HR director with reports counts as HR and a CEO with reports counts as Leader. The weekly stress check-in also has a free-text follow-up question (“What are the primary sources of stress in your work right now?”) which let us read 2,451 first-person accounts of what people are stressed about, segmented by role.

The four groups have moved differently across the three quantitative signals, and the qualitative answers from each group sound different too.

The four groups, sized

HR and Leader are small populations of roughly 1.6% of the cohort each. Managers are 15.2%. The remaining 81.6% are individual contributors.

Where each group sits today

The most recent 90 days are the cleanest cross-section. The chart below plots the share of responses in each metric's at-risk bucket for each group.

* Leader's WHO-5 cell has n=6 person-quarters in the 90-day window because the WHO-5 requires all five items answered in one quarter and Leader is a small population that responds less consistently to that instrument. Read the 33.3% as directional; the full 8-quarter Leader series (Figure 5) is more stable.

Daily happiness across 24 months

HR has the lowest mean happiness in 21 of the 24 months and the highest share of Not Great or Terrible responses in 14 of them. Leader has the highest mean happiness in every one of the 24 months, ending at 4.40 against HR's 3.93. That is a 0.47-point gap on the 1–5 scale. Most of the HR-vs-Leader happiness gap lives in the positive tail rather than the negative tail: 87% of Leader responses in the latest 90 days are Happy or Very Happy versus 73% for HR, while their Not-Great-or-Terrible shares are only 1.5 percentage points apart. Employee is the only group whose at-risk happiness share did not drift up over the window.

Weekly stress across 20 months

Leader has been the most-stressed group across the 20-month window: at least 15% of Leader weekly check-ins land on Very stressed in 19 of 20 months, with several months above 25%. The Leader share has held in roughly that band the whole time, ending essentially where it started. Manager has been the least-stressed group throughout, but the Manager at-risk share has crept up by 3.3 points over the same window. HR sits in the middle, with the highest variability of the four groups; HR's at-risk share fell 1.2 points net but moved between 7% and 22% along the way. Employee shows the largest shift: the at-risk share nearly doubled from 10.8% in Sep 2024 to 18.7% in Apr 2026, lifting Employees from the lowest at-risk share to within two points of Leader.

WHO-5 well-being across 8 quarters

* Leader has no Q1 2026 cell in this CSV; the Leader “Q1 2026” values shown are the most recent Leader quarter on file, which is Q4 2025 (n=10).

The platform-wide +6% WHO-5 rise reported in YoY Trends 2026 shows up here, distributed unevenly across groups. Manager and Employee gained the most on the composite (+3.9 and +2.9 points). Manager's share below the clinical low-wellbeing threshold dropped by 7.6 percentage points, the largest healthy shift on the at-risk side. HR's composite gained 1.5 points net, but the trajectory was not monotonic: a peak at 72.5 in Q3 2025 was followed by two quarters of decline, ending at 64.2, near HR's starting point. HR's low-wellbeing share dropped from 33.3% to 26.1%, but it remains roughly twice the Manager and Employee rates. Leader's WHO-5 mean has been the highest of the four groups throughout the window and has held near 71–72; the Leader at-risk share, with quarterly cells of n=9–11, should be read as directional. Leader has no Q1 2026 cell; the last Leader quarter shown is Q4 2025.

24-month change, at a glance

What each group is stressed about

After the weekly stress check-in, respondents are asked: “What are the primary sources of stress in your work right now?” Across 2,451 free-text responses over 24 months (51 HR, 44 Leader, 650 Manager, 1,322 Employee, after blanks and accidental responses were filtered out), the substance of the stress differs sharply by role.

HR talks about work, not people. Workload (33% of HR responses) and deadlines (28%) lead by wide margins, both the highest rates of any of the four groups. Recruiting and staffing is a distinctive HR theme at 10% (vs 1–5% for the other groups). What is striking is what is not there: family or health stress shows up in only ~10% of HR responses (versus 25% for Leader, 21% for Manager, 16% for Employee). HR is mentally consumed by their own work backlog rather than people-management or personal-life stressors.

Representative HR responses

“Just workload, with [...] additional clients and all the QBRs coming up.”

“การสรรหาคน ไม่ทันกับความต้องการของหน่วยงาน” (Recruiting people fast enough for what the business needs.)

“ความคาดหวังในตนเองและผลลัพธ์ในงานของตนเอง” (My own expectations and my own work outcomes.)

Leader talks about life as well as work. Personal/family/health is the second-largest theme at 25%, the highest of any group, and the cleanest distinguishing feature of Leader stress. Financial concerns (13.6%) is also distinctively Leader. Several Leader responses describe simultaneously running a business, supporting a parent with a serious illness, and dealing with cash-flow pressure. Workload (29.5%) is the leading theme but the texture is different from HR's. Leaders talk about “too many things to work on,” not “too many tasks I cannot finish.”

Representative Leader responses

“Lots of stress outside of work too due to health issues. But at work it's the worry that we are not making ends meet, especially for the first few months of the year where business is tight.”

“Significant family issues and challenges happening alongside numerous workplace challenges that will probably be hard to resolve.”

“Clients not paying, others delaying payments.”

Manager talks about targets and customers. Sales and revenue targets (12.2%) and customer/client demands (12.8%) are both the highest rates of any group, and team-and-people-management (6.5%) is also Manager-distinctive. Manager stress is outward-facing: the operational interface with the market and with the people they supervise. Personal/family stress is also high at 21%, second to Employee in absolute terms but a strong second concern alongside the operational themes.

Representative Manager responses

“The most challenging situation is managing low-performing team members while maintaining overall team productivity.”

“Handle client [name] as they yelling in front of my face. During UAT.”

“Revenue, profitability, personnel challenges, preparing for interview, bond issuance.”

Employee responses are the most diverse. No single theme passes 16%: personal/family/health (16%), customer demands (11%), workload (10%), deadlines (9%), financial (8%), skill-gap (7%) all compete. The distinctive Employee signal is breadth, not concentration. Employees mention learning new tools, family-finance pressures, customer escalations, project deadlines, and personal mood-and-health concerns roughly in proportion. There is no single intervention that would speak to the median Employee's stress.

Representative Employee responses

“Too many things to do, too little time, too little hands. :(”

“หลายเรื่องไม่เคยทำ และน้อยคนที่จะมีประสบการณ์กับมัน” (Many things I have never done, and few colleagues have experience with them either.)

“ปัญหาส่วนตัวของตัวเอง ทั้งสุขภาพกาย และสุขภาพใจ” (My own personal problems, both physical and mental health.)

How each group's 24 months unfolded

HR: the unhappiness outlier, slowly recovering on well-being

HR began the window with the lowest happiness mean of the four groups (3.80) and ended near where it started (3.93). The HR happiness line is the lowest of the four in 21 of 24 months. On stress, HR has been the most volatile group, fluctuating between 7% and 22% Very-stressed week-to-week and ending modestly below where it started. The WHO-5 picture is the most encouraging signal for HR: the low-wellbeing share fell from 33.3% in Q2 2024 to 26.1% in Q1 2026, a meaningful net move, but the trajectory peaked in mid-2025 and slid back, ending two quarters below its high. HR remains the only group with more than 1 in 4 below the clinical low-wellbeing threshold today. The qualitative picture is consistent with the quantitative one: HR talks about volume and capacity, not about acute distress.

Leader: high engagement and high pressure, holding steady at both extremes

Leader has been the happiest group every month in the 24-month window (mean 4.40 in the latest month, vs 3.93 for HR and 4.01 for Manager), and simultaneously the most stressed (mean 1.73, with 26% of weekly check-ins reporting Very stressed). Neither pattern has moved much: mean stress eased slightly over 20 months (-0.13), but the at-risk Very-stressed share held flat at the top of the chart (20.8% → 20.9%). Mean WHO-5 stayed near 71–72 throughout. Stress and engagement are co-located in this group and have been the entire time. The qualitative answers add a layer the quantitative signals cannot: a substantial fraction of Leader stress is personal/family/health and financial, not just work volume.

Manager: healthiest today, but the trend lines bear watching

In the most recent 90 days, Manager has the lowest stress (mean 1.46, 11.0% Very stressed) and the lowest WHO-5 at-risk share (10.2%) of the four groups. The 24-month trajectory was a meaningful improvement on WHO-5 (composite +3.9; at-risk share −7.6 percentage points, the largest healthy shift of any group on any metric) and a clear deterioration on stress (mean +0.16; at-risk share +3.3pp). Happiness was essentially flat. Managers ended the window in better shape on well-being and worse shape on stress. What they say they are stressed about is targets and customers, with team-management a close third.

Employee: the at-risk stress shift is the biggest movement in the study

Daily happiness and quarterly WHO-5 look largely stable for Employees, with small improvements on both lenses (mean happiness +0.07, mean WHO-5 +2.9, at-risk WHO-5 essentially flat). The stress signal moved much more: the share of weekly check-ins reporting Very stressed nearly doubled, from 10.8% in Sep 2024 to 18.7% in Apr 2026. That is the largest at-risk-share shift of any group on any signal in this study, and it lifted Employees from the lowest at-risk stress share to within two points of Leader. The qualitative answers do not point to a single cause: Employee stress sources are the most diverse of any group, with no theme above 16%.

What this means for each group

For HR teams

The most stable finding in this study is that HR sits at the bottom of the happiness scale and the bottom of the WHO-5 scale, consistently, for two years. HR is not the most-stressed group; their stress profile is in the middle of the pack. That pattern is consistent with persistent, low-grade dissatisfaction rather than acute pressure. The HR happiness mode is “Okay” (23.7% of HR responses) rather than “Happy” or “Very Happy.” The qualitative answers reinforce this: HR talks about workload and deadlines, not about people problems or personal crises. That pattern fits the “cobbler's children” reading from the prior HR eNPS Bellwether study: HR teams build well-being programmes for others and do not benefit from them themselves. Programmes targeted at HR may want to reach for capacity relief and engagement rather than crisis-tier support, since what HR describes is being overworked rather than being in distress.

For Leaders

Leader is the most-engaged and most-pressured group, and those two properties have travelled together unchanged for 24 months. That stability matters: the 20% Very-stressed share is a structural feature of the role in this data, not a transient spike that a quarter of relief will resolve. Leaders also have the highest mean happiness of the four groups, which is the signal that this is engagement-with-cost rather than burnout. The free-text answers add an important piece: 25% of Leader stress is personal/family/health, and 14% is financial. Both shares are higher than for any other group, and they often co-occur in the same response (aging parents, health crises, cash-flow pressure). The practical implication is that “leader well-being” programmes that only address work-stress will likely be mis-targeted; a real Leader recovery offer needs to make space for life events outside work, not only optimize the meeting calendar.

For Managers

Manager looks the healthiest of the four groups in the current snapshot: lowest stress, lowest WHO-5 at-risk share, and the largest WHO-5 recovery over the window. Reading that as “managers are fine” would miss the more important signal. Stress at-risk share rose 3.3 points over 20 months, and what Managers say they are stressed about is operational: sales targets, customer demands, and team-performance issues. Those are the kinds of pressures that compound if revenue conditions tighten further. The gap between Manager and Employee at-risk stress is closing from below. The conventional “manager burnout” narrative is not visible in the current snapshot. It would be the first place to look if a future snapshot turned worse, and the topics Managers name as stressors are the topics most likely to move that signal.

For Employees

Employees are 82% of the cohort and the group whose trajectory matters most for any company-wide well-being claim. Happiness and WHO-5 mean look stable for them, but the at-risk stress share nearly doubled in 20 months. That is the largest categorical move in the study. The qualitative answers do not point to one cause: Employee stress sources are the broadest of the four groups, with personal/family, customer demands, workload, deadlines, financial, and skill-gap concerns all competing in roughly equal proportion. Two readings are consistent with this: either the rising stress is the workforce-wide expression of the engagement collapse reported in the prior YoY Trends study, or it reflects a structural change in how work is being absorbed by the largest population. Either reading suggests that company-wide stress mitigation is warranted (this is not a leadership-only or HR-only issue), and that a one-theme intervention will under-serve most Employees, because no single dominant theme is there to address.

Limitations

• Observational only. Differences between groups may reflect who chooses to enter a role as much as what the role does. No causal claim is made.
• Group priority is a choice. The mutually-exclusive ordering HR > Leader > Manager > Employee places an HR director with reports in HR. A different priority would give different group means and shares.
• Leader definition. The Leader group is restricted to people at the top of their company's org chart who also have at least one direct report. The “has reports” requirement is what separates true top-of-org leaders from employees whose reporting chain is simply missing or out of date in the underlying records.
• Voluntary response. All instruments are opt-in. The direction of any response bias on low- vs high-wellbeing respondents is not known.
• Small Leader/HR WHO-5 cells. Quarterly WHO-5 cells for these groups range n=6 to n=28. Treat quarterly Leader values as directional.
• Stress series is 20 months, not 24. The weekly stress survey rolled out in Sep 2024; the first months of the series come from a smaller stress-active cohort than the last months.
• Qualitative themes are pattern-tagged, not LLM-coded. Each response is matched against a fixed set of theme patterns in English and Thai. Coverage is uneven; roughly 30–47% of responses do not match any of the named themes and fall into “Other”, depending on the group. The cross-group ordering of themes is robust; the absolute share for any one theme is a lower bound.
• No company-level random effects. Means and shares pool across 117 companies; a single large company could move a group's number.