Here is the most expensive contradiction in remote management. Microsoft surveyed 20,006 knowledge workers and found that 85% of leaders say the shift to hybrid work makes it hard to be confident their people are productive — while 87% of employees say they already are. Microsoft calls that gap "productivity paranoia." Only 12% of leaders report full confidence in their team's output.
The reflex has been to start watching. A 2025 ExpressVPN survey found that 74% of U.S. employers now use online tracking tools, and 61% run AI-powered analytics to score productivity. Most added the software recently.
There is a century-old experiment that predicts exactly how this ends. In the 1920s, researchers at a factory outside Chicago found that people work differently the moment they know they're being observed. They named it the Hawthorne effect — and it is the single best lens for understanding why surveilling a distributed team produces a short, hollow bump and then quietly backfires. This piece breaks down what the studies really found (including the part that debunks them), how the Hawthorne effect in remote work splits into two very different kinds of being watched, why surveillance backfires, and how to get the productive version without a single tracking tool.
What the Hawthorne Effect Actually Found
The name comes from Western Electric's Hawthorne Works in Cicero, Illinois, where studies ran between 1924 and 1932. The most famous were the illumination experiments: researchers raised the lighting and output went up — then they lowered it and output went up again. The lighting wasn't the variable. The workers got faster because they knew they were being studied. As Encyclopaedia Britannica puts it, under the team led by Elton Mayo, "job performance improved because more attention was being paid to the workers."
Then comes the part most managers never hear. The lift didn't last. Once the attention stopped, productivity "returned to their original level" — the gains "disappeared when the attention faded." The Hawthorne effect, in its original form, is a temporary response to novelty, not a permanent gear you can leave the team in.
It's even shakier than the legend suggests. In 2011, economists Steven Levitt and John List recovered the original illumination data and concluded that "existing descriptions of supposedly remarkable data patterns prove to be entirely fictional," though they did find "more subtle manifestations of possible Hawthorne effects." Translation: the effect is real but small and fragile. So if you're betting your remote-management strategy on the idea that watching people will lift their output, you're betting on the weakest, most short-lived effect in the management canon.
The Hawthorne Effect in Remote Work: Two Ways to Be Watched
The Hawthorne effect — a workplace form of the observer effect, where being watched changes what you're trying to measure — has a good twin and an evil twin, and the difference is everything.
Being seen — noticed, valued, given attention — is the version Mayo documented. It signals your work matters and someone is paying attention to it. Being surveilled — keystroke logs, screenshots on a timer, an activity score — sends the opposite signal: we don't trust you to work unless we're staring. In an office, the difference was obvious from body language and proximity. Remote work flattened both into the same glowing status dot, so leaders reach for the only "attention" a dashboard can deliver: monitoring.
That's why the bossware boom and productivity paranoia are the same story. When 74% of employers track activity and 59% use real-time screen capture, they're trying to manufacture the Hawthorne effect at scale. But you can't fake the good twin with the evil one. Constant, involuntary observation doesn't read as you matter; it reads as suspicion — which is why the trust crisis around recorded and monitored work has become a defining issue for distributed teams, and why remote team trust is now the metric that actually predicts performance. The Hawthorne effect in remote work tells you what surveillance really produces: behavior tuned to the watching, not to the work.
Why Employee Monitoring Backfires
Watching harder doesn't close the confidence gap. It widens it, in three predictable ways.
It manufactures the wrong behavior. A study by Chase Thiel, David Welsh, and colleagues, written up in Harvard Business Review, found that monitored employees were substantially more likely to break rules — taking unapproved breaks, ignoring instructions, even stealing equipment. The mechanism is the unsettling part: monitoring strips people of their sense of agency, so they feel less personally responsible for their own conduct. Surveillance doesn't just fail to create accountability. It actively erodes it.
It costs you trust and people. A February 2025 MIT Technology Review feature put it bluntly: "monitoring greatly increases worker stress and can break down trust between an employer and its workforce," and "roughly half [of tech workers] would rather quit than be monitored." The survey data agrees — 49% of employees say they'd consider leaving if surveillance increased, and 24% would even take a pay cut to avoid invasive monitoring. You don't retain your best remote performers by treating them like suspects.
It expires anyway. Even the generous reading of the Hawthorne effect says the bump fades when the novelty does. Surveillance is permanent attention of the worst kind, so you skip the temporary lift and go straight to the decay — now with resentment attached. This is the principal-agent problem at work in its purest form: spending real money on oversight that makes the underlying misalignment worse, not better.
Productivity Theater Is What You Actually Buy
When the metric is "looks busy," people get very good at looking busy. Visier found that 43% of U.S. employees spend more than 10 hours a week on "productivity theater," and 83% admit to at least one performative work behavior. That's more than a full day a week spent performing work instead of doing it.
Monitoring makes it worse, because it gives people a clear target to game. The same ExpressVPN research found employees keeping unnecessary apps open (16%), scheduling emails to appear active (15%), and using mouse jigglers or virtual backgrounds to evade detection (12%). This is Goodhart's law in motion: the moment activity becomes the target, activity stops measuring anything real. You wanted output and you bought a green dot. Worse, the theater displaces the deep, uninterrupted work hours that actually move the business — the exact thing the dashboard was supposed to protect.
How to Trigger the Good Hawthorne Effect Without Surveillance
The fix isn't to stop paying attention — attention is the active ingredient. It's to change what you observe and how. Here's the playbook for managing remote teams that gets the good twin of the Hawthorne effect on purpose.
Watch outcomes, not activity
The most durable form of observation is the one aimed at results. In a results-only or outcome-based model — the philosophy behind GitLab's all-remote handbook — leaders set the destination, document what "done" looks like, and judge the work, not the clock. Output observation doesn't decay the way novelty does, because shipped work is real. A manager who reviews the deliverable every week is paying attention to exactly the thing that matters, and the team responds to it.
Make work visible by default, not on demand
There's a sharp line between surveillance and transparency. Surveillance is a manager pulling activity logs behind your back. Transparency is work that's public by default — shared docs, async updates, a canvas anyone can open. One feels like being watched; the other feels like being seen. Building a distributed team around default visibility means progress is legible to everyone without a single screenshot, and it kills the "what is my team even doing?" anxiety at the source.
Replace the watch with genuine attention
Mayo's finding was that people respond to being noticed and valued — so do that deliberately. Recognition, real one-on-ones, and going to see the actual work the constructive way. A remote gemba walk, where a leader observes the real work to help and unblock rather than to catch people out, is the positive Hawthorne effect by design. The observation is identical; the intent is the opposite, and people can always tell the difference.
Build shared context so you don't need to watch
Most surveillance is really a proxy for I have no idea what's happening on my team. Fix that visibility gap directly and the urge to monitor evaporates. When the work itself — the meeting, the decisions, the whiteboard — is captured and searchable, you can see where things stand without watching anyone. That's the bet behind tools like Coommit: video, an interactive canvas, and a contextual AI in a single session, so the output and the why behind it are visible by default instead of reconstructed from activity logs. You're observing the work, which is exactly where attention belongs.
Measure Output, Not Eyeballs
Swap the activity dashboard for the things that compound: work shipped, decisions made, blockers cleared, customer outcomes moved. If you genuinely can't tell whether a remote teammate is productive without screen recordings, the real problem isn't a missing tracker — it's fuzzy goals and unclear ownership. Tighten those, and "are they working?" answers itself, because the work is right there in the open. Clarity is the accountability tool that surveillance only pretends to be.
Conclusion
The Hawthorne effect in remote work is a warning and an instruction in the same breath. Attention changes behavior — so spend it on the work, not on the worker. Surveillance buys you productivity theater, an exodus risk, and a bump that fades the moment you blink. Visible outcomes and genuine attention buy you the real thing, and they keep paying out long after the novelty wears off.
The teams that win the next phase of remote work won't be the ones with the best tracking software. They'll be the ones who made the work itself so visible that watching the worker became pointless. Build a workspace where the meeting, the canvas, and the decisions are captured and shared by default — like the one Coommit is built around — and the dashboard you actually need is the work, out in the open, for everyone who's paying the right kind of attention.