How to Monitor Employees Working From Home Without Wrecking Trust
Remote work monitoring is the practice of collecting records of how distributed employees use company systems, so an employer can answer a question it cannot answer by walking across an office. If you are searching for how to monitor employees working from home, the useful answer starts one step earlier than the tool: in almost every case the thing you actually want is visibility of the work, and activity surveillance of the person is a poor substitute that costs more than it returns.
We are an independent publication. We sell nothing in this category, we take no referral fees, and we have not tested any product, so there are no names, prices, or feature claims below. That constraint is also the point. A page funded by commissions on monitoring signups cannot comfortably tell you not to buy one, and for most of the managers who reach this question, not buying one is the right call.
What follows is the case for measuring the work instead, the four problems that exist only when the office is a home, what to do if you genuinely do need device-level records, and what US law expects before any of it starts.
The honest answer to how to monitor employees working from home
Measure the work, not the person. Nearly every request we see for remote monitoring is really a request for one of three things: confidence that commitments are being met, early warning when something is going wrong, or evidence for a decision someone has already half made. Activity data is bad at the first two and dangerous for the third, because it produces numbers precise enough to feel like proof and vague enough to support any story. Notice duties follow the employee, so check employee monitoring laws by state for wherever your team actually sits.
The substitute is unglamorous. Agree what each person owns and when it is due. Put the state of that work somewhere visible so nobody has to ask. Review the gap between what was promised and what landed, weekly, in public. This takes a manager's attention rather than a purchase order, which is precisely why the purchase order is tempting. The rules themselves belong in a written remote work policy before any of this is switched on.
There is one honest limitation to that position. If a manager has fifteen direct reports across four time zones, telling them to pay closer attention is not advice, it is a criticism. Fixing span of control and reporting cadence is the real intervention, and no tool substitutes for it. Our guide to how to measure employee productivity when you cannot see the work covers what to put in place instead.
What changes when the office is someone's home
Three things change: the boundary, the device, and the clock. In an office, the boundary of observation is a building the employer owns and everyone can see the edges of. At home, the same collection reaches into a room with a family in it, and the employee has no way to step outside the observed zone without stopping work. That asymmetry is the source of most of the resentment, and it is not solved by a friendlier dashboard.
The distribution is settled enough that this is now the normal case, not an edge case. The Pew Research Center published figures in January 2025, fielded October 7 to 13, 2024 among 2,315 US workers whose job can be done from home, showing 75% work remotely at least some of the time. The base matters: that describes remote-capable workers, not the entire US workforce. The same survey found 75% are now required in the office certain days, up from 63% in early 2023, and that among hybrid workers, 72% prefer hybrid.
The clock is the third change, and the most underrated. Work at home spreads into the evening without anyone deciding it should, and an activity log will record that spread as commitment rather than as a warning. Reading those late signals correctly is a skill, and we wrote separately about the signs of employee burnout that show up in operational data before anyone says anything out loud.
Four problems that only exist at home
Four home-specific problems break monitoring designs that worked fine in an office. Each one has a response that is cheaper and less invasive than the monitoring feature usually proposed to solve it.
| Problem | What goes wrong | Why monitoring makes it worse | What to do instead |
|---|---|---|---|
| Personal devices | Work happens on a laptop the company does not own | An agent on a family machine collects household activity you cannot lawfully justify or safely store | Issue company equipment for anything needing oversight, or secure the application layer instead of the device |
| Household members in range | Cameras and screen capture pick up partners, children, and roommates | You are recording people who never entered any agreement with you | No standing camera requirement, no continuous screen capture at home |
| Blurred hours | The day fragments around caregiving, errands, and evening catch-up | Idle time gets read as absence, and late activity gets read as dedication | Define core overlap hours and judge delivery against agreed dates |
| Time zones | Colleagues are online at different times by design | Presence dashboards make everyone look absent to someone | Default to asynchronous handoffs and measure whether work moves, not when |
The clock problems are visible in independent data. Microsoft's Work Trend Index, published June 17, 2025 and combining anonymized Microsoft 365 telemetry with a survey of 31,000 knowledge workers across 31 markets fielded February 6 to March 24, 2025, reported that meetings after 8 pm rose 16% year over year, that 29% of people are still checking the inbox by 10 pm, that 40% of those online at 6 am are triaging email, and that 30% of meetings span multiple time zones. An activity graph drawn over that pattern rewards exactly the behavior you should be trying to reduce.
How to monitor employees working from home if you genuinely must
Some situations require device-level records, and pretending otherwise helps nobody. Regulated recording obligations, security incident response, controlled data handling, and roles with mandatory audit trails are real. If you are in one of those, the question is not whether to collect but how narrowly. Here is the sequence we would defend. The product names in this space are unhelpfully similar, and our guide to remote computer monitoring software separates remote access from endpoint management from activity capture.
- Write the question you are trying to answer in one sentence, and get it agreed before any tool is discussed.
- Check whether your identity provider, document storage, or ticketing system already answers it. Very often they do.
- Restrict collection to company-owned equipment. Never install an agent on a personal machine.
- Choose the narrowest fields that answer the question, and write down the fields you are refusing to collect.
- Rule out continuous screen capture, webcam access, microphone access, and keystroke content in a home setting.
- Give written notice, get it acknowledged, and post it, before collection starts rather than after.
- Report at team level by default. Require a written reason to view a named individual, and log every lookup.
- Set automatic deletion per field, and a review date where the program must show a decision it changed or be switched off.
Two of those steps do most of the work. Restricting collection to company equipment keeps the household out of your systems, and setting a review date stops the program outliving its reason. If you want the technical detail behind what each field actually captures, we explain monitoring a computer at the operating system level in a separate piece.
Where this advice does not apply
If you run a contact center under recording obligations, handle regulated financial or medical data, or operate a security function that must reconstruct incidents, treat the skeptical framing here as background rather than instruction. Your monitoring is not discretionary and the useful questions are about scope, access control, retention, and disclosure. The argument in this article is aimed at the much larger group whose real problem is that a manager cannot tell what is happening, which is a reporting problem wearing a surveillance costume.
What to measure instead of activity
Measure commitments and flow. For each person or team, the useful record is what they said they would deliver, whether it landed on the agreed date, and how long the work sat waiting on someone else. Add one or two quality signals a customer would recognize, such as rework or escalations. None of that requires an agent on a laptop, and all of it survives a conversation with the person it describes.
Hours still matter for some work, and pretending they do not is its own kind of dishonesty. Non-exempt roles, client billing, and project costing all need a record of time. The distinction is between a time record the employee creates and can see, and a surveillance feed the employee cannot inspect. Our guide to time tracking for remote employees covers how to run the first without sliding into the second.
There is evidence that the arrangement itself can be made to work without heavy oversight. A randomized controlled trial reported in Nature in June 2024 by Bloom, Han and Liang, covering 1,612 employees over six months and randomizing them to five days in the office versus three days in the office with two at home, found quit rates about a third lower in the hybrid group, with no measured impact on performance reviews or promotions and higher job satisfaction. That is a study of the schedule, not of monitoring, and it is worth keeping the distinction clean. For the wider picture we maintain a page of verified remote work productivity figures with their samples and fielding dates attached.
What US law expects from remote work monitoring
Remote work multiplies your legal exposure, because notice duties follow the employee's location rather than the company's. A single monitoring configuration applied to a distributed team can be compliant in one state and not in another, which is why the practical standard is the strictest state you employ in, confirmed with counsel.
New York Civil Rights Law section 52-c is the clearest example. Employers who monitor telephone conversations, email, or internet usage by an electronic device must give prior written notice upon hiring to all employees subject to monitoring. The notice must be in writing or electronic form and acknowledged by the employee, and it must be posted conspicuously where affected employees can see it. Penalties run up to $500 for a first offense, $1,000 for a second, and $3,000 for a third and each subsequent offense, with an exemption for processes managing the type or volume of email, voicemail, or internet usage performed solely for computer system maintenance or protection. The statute text is worth reading directly.
We describe New York because it is checkable, not because it is universal. Monitoring law varies by state, and the federal Electronic Communications Privacy Act sits behind any interception question. If you employ staff in the EU, GDPR adds requirements beyond anything covered here. Separately from the law, there are categories of data that a home setting should place out of bounds regardless of what is technically permitted, and we set those out in our page on what employee monitoring should never collect.
What getting this wrong actually costs
The cost is people, and it is measurable in advance. Pew's January 2025 figures found 46% of remote-capable workers would be unlikely to stay in their job if they could no longer work from home, with 26% saying very unlikely. That is a workforce that has already demonstrated it will change jobs over working conditions, and a badly introduced monitoring program is a change to working conditions whatever the announcement calls it. The same arithmetic applies to staff monitoring programs generally.
Layer on the expectation people bring. Pew found in April 2023 that 81% of US adults believe workers would feel inappropriately watched if employers used AI to collect and analyze information about how they do their jobs, 52% saying definitely and 29% probably. Your announcement does not land on a neutral audience. It lands on one that has already decided how this usually goes, and the burden of proof is yours from the first sentence.
So the calculation is not monitoring versus nothing. It is a small amount of extra confidence, weighed against a retention risk in a population with options, in exchange for data that mostly describes interaction with a machine. When we run that sum for the teams who ask us how to monitor employees working from home, it comes out negative more often than not.
Key takeaways
- Measure the work, not the person. Commitments, dates, and flow answer the real question better than activity data does.
- Four problems are home-specific: personal devices, household members in range, blurred hours, and time zones. Each has a cheaper answer than a monitoring feature.
- Microsoft's June 2025 research found meetings after 8 pm up 16% year over year and 30% of meetings spanning time zones, which is why presence dashboards mislead for remote teams.
- A randomized trial reported in Nature in June 2024, covering 1,612 employees, found quit rates about a third lower under hybrid working with no measured impact on performance reviews.
- Pew found in January 2025 that 46% of remote-capable US workers would be unlikely to stay without work from home, 26% very unlikely. That is your downside.
- Notice duties follow the employee's state. New York requires prior written notice, acknowledged and posted, with penalties up to $3,000 per offense. Confirm your obligations with counsel.
Frequently asked questions
How do you monitor employees working from home?
The version that works is not surveillance of the person, it is visibility of the work. Agree what each person owns, when it is due, and where progress is visible without asking. If you still need device-level records for a security or compliance reason, keep the collection on company equipment only, restrict it to the narrowest fields that answer your stated question, tell people in writing before it starts, and put a deletion schedule on everything you keep.
Is it legal to monitor employees working from home in the US?
Monitoring company equipment is broadly permitted in the United States, and notice duties vary by state. New York Civil Rights Law section 52-c requires employers who monitor telephone conversations, email, or internet usage by electronic device to give prior written notice upon hiring to all employees subject to monitoring, acknowledged by the employee and posted conspicuously, with penalties up to 500 dollars for a first offense, 1,000 for a second, and 3,000 for a third and each one after. Remote teams create obligations in every state where someone sits, so confirm with counsel.
Can an employer monitor a personal computer at home?
Only if the employee installs something, and we think most employers should refuse to ask. Putting an agent on a device the company does not own extends collection into a household, mixes family data into a corporate system, and creates a retention problem with no clean ending. The better answer is to issue company equipment for work that needs oversight, or to secure the application layer so the personal device never holds the sensitive material in the first place.
Should remote employees have to keep their webcam on?
No, not as a standing requirement. A camera in a home points at a bedroom, a kitchen, a partner, or a child, and none of that is the employer's business. Always-on video also produces nothing usable: presence in front of a lens is not work. Ask for cameras in specific meetings where they help the conversation, make it a request rather than a rule, and never record continuously.
What should you measure instead of activity for remote workers?
Measure commitments and cycle time. What did this person or team say they would deliver, did it land, and how long did the work sit waiting on someone else. Add a small number of quality signals your customers would recognize, such as rework rate or escalations. These are harder to collect than a keystroke count and vastly more useful, because they describe the outcome you actually care about rather than the interaction with a machine.
How do you handle time zones without tracking hours?
Publish a narrow band of overlap hours and treat everything outside it as asynchronous by default. Microsoft reported in June 2025 that 30 percent of meetings span multiple time zones, which is exactly the condition that turns an activity dashboard into noise, because someone will always look idle to someone else. Judge the handoff, not the hour. If work moves cleanly between regions, the schedule is working.
Does monitoring make remote employees more productive?
We have no verified evidence that it does, and we would not repeat the claim without it. What is documented is that hybrid arrangements can hold performance steady while improving retention. A randomized controlled trial reported in Nature in June 2024, covering 1,612 employees over six months, found quit rates about a third lower for employees assigned to a hybrid schedule with no measured impact on performance reviews or promotions. That is evidence about the arrangement, not about surveillance.
What are the risks of monitoring remote employees?
Capturing household data you never wanted, creating records you have to produce in litigation, measuring presence instead of output, and losing people. Pew Research Center data published in January 2025, fielded October 7 to 13, 2024 among 2,315 US workers whose job can be done from home, found 46 percent would be unlikely to stay if they could no longer work from home, with 26 percent very unlikely. That is a workforce already willing to move over working conditions.
How do you tell if a remote employee is struggling without monitoring them?
Watch the shape of the work rather than the person. Commitments slipping without being renegotiated, work in progress climbing, questions drying up, and messages arriving at hours nobody should be working are all visible in tools you already run. Then have the conversation. A pattern is a reason to ask how someone is doing, and it is never a substitute for asking.
When is remote work monitoring genuinely justified?
When a specific obligation requires it. Regulated environments with recording duties, security incident response, handling of controlled data, and roles with mandatory audit trails all sit outside the general argument. In those cases the question is not whether to monitor but how narrowly, who may look, how long records live, and how clearly it is disclosed. Everything else should have to justify itself against a plain alternative: clearer expectations and better reporting.