When disconnection follows a role, not a person, the cause is structural—not individual.

Whether people feel safe speaking up depends more on their manager than on company policy.

Jumping to individual support first risks telling employees the problem is them.

This post is co-authored by Nathaniel Sabater, Mariangel Maldonado, and Hans Rocha IJzerman.

A company downsizes, survives the chaos, and launches a wellness initiative—mental health days, resilience workshops, a new app. Utilization stays low. Burnout stays high. Leadership is baffled.

This isn't a story about bad intentions. It's a story about skipping the diagnosis.

Workplace loneliness is one of the most expensive and least understood problems in organizational life. Yet most companies respond to it with individual-level interventions—therapy access, mindfulness tools, mental health leave—regardless of where the problem actually originates. When the root cause is structural or managerial, individual supports don't just underperform. They quietly send struggling employees the message that the problem is them.

Getting the intervention right means getting the diagnosis right first. And that diagnosis has to happen at three distinct levels.

Level 1: The Organization Itself

The first question to ask when disconnection is widespread: Have we designed conditions that make connection structurally impossible?

This is more common than it sounds. Chronic overload, hybrid policies that exist on paper but not in practice, blurry roles after a restructuring, calendar density that leaves no slack for the informal exchanges where connection actually forms—these don't just exhaust people, they eliminate the moments that matter. The Job Demands-Resources model has long documented how resource depletion undermines employee well-being, but it's still underused as a diagnostic lens.

One reliable signal: The same disconnection patterns show up across different teams, different managers, different people—all holding the same role. When the problem follows the role rather than the person, you're looking at the structure.

The mismatch: offering individual support before fixing impossible workloads. People take the mental health day and return to the same conditions. The root cause stays intact.

Once structural factors are ruled out, the most powerful variable is often the one least formally measured: what kind of climate the manager actually creates.

Two teams inside the same organization—same policies, same resources, same office—can produce completely different levels of belonging. That's not mysterious. We know from decades of psychological safety research that whether people feel safe enough to speak up, ask for help, or admit they're struggling depends far less on company policy than on what the person in charge models and tolerates day to day.

When employees repeatedly turn in below-expectation work rather than asking a clarifying question, that's almost never a process problem. It's a signal that asking felt risky. You can redesign the briefing template; it won't touch the underlying dynamic.

The mismatch: improving systems and workflows when the real constraint is that people don't trust the environment enough to use them honestly.

Level 3: The Individual

Individual-level support—coaching, counseling, tailored accommodations—matters. But it should be the last stop in the diagnostic process, not the first.

Organizations jump here because it's visible. You can point to the program, count sign-ups, and report utilization to the board. But when structural and team-level problems get reframed as individual struggles, the implicit message is corrosive: You're not adapting well enough to conditions we haven't actually fixed.

There are genuine individual factors that shape workplace loneliness—caregiving demands, life transitions, neurodiversity, mental health. These deserve serious, flexible support. But one of the most consistent mismatches I see involves connection events scheduled after hours, framed as optional but culturally loaded, that systematically exclude the employees who most need support. Research on how isolation spreads across relational contexts helps explain why: When work schedules crowd out community and family life, the disconnection doesn't stay at work. It compounds. What was designed as a connection program becomes a new source of exclusion.

Why Diagnosis Is Hard

Standard engagement surveys tell you that a team scores low on belonging. They rarely tell you why—whether it's an overloaded structure, an unsafe manager, or someone navigating a caregiving crisis. All three look the same on a five-point scale.

What's needed is diagnostic sequencing. Structural problems show up in human resources information system data and workload patterns. Team climate emerges from manager feedback and from differences across teams operating under identical organizational conditions. Individual factors require longitudinal, person-specific observation—the kind of data that population averages erase.

The sequence matters because mismatched interventions don't just waste budget. They erode trust. They signal that leadership doesn't actually understand the problem. And they leave the root cause intact while people quietly start looking for the exit. There's decent evidence that supporting community engagement—volunteering, civic participation—improves employee health and workplace cohesion in ways that internal programs alone can't replicate, which is part of why organizations that invest only inward tend to plateau.

What This Means in Practice

Organizations making real progress on workplace loneliness are not necessarily the ones with the biggest wellness budgets. They're the ones willing to slow down before launching anything and ask: At what level does this problem actually live?

Mental health days can't fix impossible workloads. Team-building events can't repair psychological safety gaps created by managers. Individual accommodations, however thoughtful, can't compensate for environments structurally designed to exhaust people.

Gather the right evidence first. The right intervention becomes a lot less elusive.

Nathaniel Sabater is with Annecy Behavioral Science Lab, and Mariangel Maldonado is former head of well-being at Booking.com. A version of this post also appears on the Annecy Behavioral Science Lab website. It is a follow-up to our earlier Psychology Today post "Before Your Organization's Next Connection Program: Diagnose."

There was a problem adding your email address. Please try again.

By submitting your information you agree to the Psychology Today Terms & Conditions and Privacy Policy

© Psychology Today