How to know if remote employees are working, without spying on them
The most reliable way to know if remote employees are working is to set clear outcomes and check delivery against them, then use work data to spot patterns like overload or long idle stretches. Signals such as mouse movement are easy to fake; patterns across apps, output and availability are not.
This guide is for managers and HR teams who need to answer that question fairly. It covers the signs worth noticing and what else they can mean, which data helps, what to do about mouse jigglers and second jobs, how to monitor remote employees without hurting trust, and a fair process for when you suspect a problem. For the technical side, how to view activity on a work computer, see how to see what employees are doing on their computers.
Table of Contents
- How do you know if a remote employee is working?
- What signs suggest someone isn't working, and what else could they mean?
- Which data helps, and which misleads?
- Can you tell if someone uses a mouse jiggler?
- What if a remote employee has a second job?
- How do you monitor remote employees without hurting trust?
- What should you do if you suspect someone isn't working?
- Frequently Asked Questions
How do you know if a remote employee is working?
Start with outcomes: agree what each person should deliver and by when, then check delivery against it. Work patterns come second, to explain gaps and spot overload early, not to prove that someone looked busy.
In an office, managers noticed who was at their desk and who left early. Those signals were always unreliable, but they felt like oversight. Remote work removes them, and office habits don't translate well: remote knowledge work happens in less visible cycles of deep thinking, writing, research and asynchronous collaboration. The same applies when you compare remote and in-office work; see hybrid work management and return-to-office analytics.
A workable routine has four parts:
- Agree outcomes for each role. Deliverables and deadlines, quality measures such as client feedback or revisions, and contribution to team goals. Roles differ: developers need long focus blocks, sales reps need communication volume, support staff need fast responses. For engineering teams, see developer and IT team productivity software. For outcome and time measures that fit each kind of role, see how to measure employee productivity.
- Agree working norms. Core hours or availability windows, expected response times, and how people show they are in focus time or away. Write them down, so "available" means the same thing to everyone.
- Review progress on a rhythm. Short written updates on what got done, what's next and what's blocking, plus a regular one-to-one. Reply to updates quickly, so they don't turn into paperwork nobody reads.
- Watch patterns, not moments. Leading indicators such as time worked against expected hours, time in core work tools and after-hours work show trouble before a deadline slips. Look at weeks, not a single afternoon.
Outcome checks on their own only reveal a problem after a deadline is missed, and patterns on their own can't tell you whether the work was any good. Together they answer how to know if remote employees are working without watching anyone's screen.
What signs suggest someone isn't working, and what else could they mean?
Missed deadlines, slow replies and gaps in availability are the usual signs, but each has other explanations: overload, unclear priorities, blockers, health or caring responsibilities. Treat a sign as a reason to ask, not as an answer.
| Sign | What it could mean | What to check |
|---|---|---|
| Missed deadlines or slipping commitments | Too much work, unclear priorities, a blocker on someone else's side, or disengagement | Workload compared with the rest of the team, whether the priority was clear, dependencies |
| Slow replies during agreed hours | Focus time, meetings, a different time zone, or absence | Whether response times were agreed, the calendar, the time zone |
| Gaps in availability during agreed hours | Work away from the computer (calls, reading, thinking), unrecorded time off, or not working | Agreed hours, recorded time off, and whether the gap repeats over weeks |
| Long idle stretches | Meetings without the computer, a computer left on after the day ended, or long breaks | Idle time settings and the person's own explanation |
| Very long hours or frequent after-hours work | Overload or burnout risk, not slacking | How work is spread across the team, and the trend over several weeks |
| A sudden change in someone's pattern | Personal circumstances, a new project or tool, or disengagement | What changed in their work, and a conversation |
Two patterns deserve particular care. Low output with long hours usually points to overload or unclear priorities; see the signs of team burnout and how to rebalance work with workforce optimization. And when paid hours regularly go to something other than work, that's employee time theft, which calls for a written policy and the same fair process described below.
Some things aren't signs at all: a camera off in meetings, a quiet presence in chat, or working at different times from you. Quiet people can be highly productive.
Which data helps, and which misleads?
Data about outcomes, and about patterns over several weeks, helps. Single input signals mislead: keystroke counts, mouse movement and screenshots show that something happened on a computer, not that useful work got done.
The biggest mistake is equating activity with output. A developer thinking through an architecture problem while staring out the window is working. A marketing manager reading industry reports on their phone is working. An employee who types 10,000 keystrokes while copying and pasting between documents may not be doing anything useful at all.
| Data | What it tells you | Where it misleads |
|---|---|---|
| Deliverables against commitments | Whether the work is getting done | It lags: you find out after a deadline |
| Time worked against expected hours | Whether agreed hours are covered | It misses work away from the computer |
| Time in core work tools | Where the working day goes | It depends on tools being categorized correctly |
| Share of time with keyboard or mouse input | How much physical input there was | It is low for reading, calls and thinking, and easy to fake |
| Keystroke counts | Typing volume | It measures typing speed, not work |
| Screenshots | What was on screen at one moment | They capture personal information, take time to review and feel like surveillance |
| Online status or hours logged | Presence | It rewards being online, not results |
Activity is the clearest example. In Intelogos, the Activity KPI is "the level of interaction with the computer through mouse and keyboard usage, expressed as the percentage of active seconds out of the total number of seconds tracked." For most jobs, 20 to 50% is typical and healthy, and it's nearly impossible to reach 100% (Activity KPI). A low number on a day of calls and reading is normal, and a manager who judges people on activity alone gets worse data and more gaming.
Always pair numbers with a conversation. Dips have reasons, such as personal issues, unclear requirements, dependency bottlenecks or tool problems: numbers show where to look, not why. For why screen capture isn't needed to understand work, see employee monitoring without screenshots.
Can you tell if someone uses a mouse jiggler?
Often, yes. A mouse jiggler, a device or app that moves the pointer, produces steady input while nothing else changes. But jigglers are a symptom of measuring activity instead of work, and the lasting fix is to change what you measure.
How mouse jiggler detection works
Real work is uneven. People switch between applications, documents and messages, type in bursts and pause to read. A jiggler creates input that is regular and continuous for long stretches while the same application and the same document stay in front, with no switching, no typing and nothing produced. Input that never pauses across a whole day is also very hard for a person to sustain: the Intelogos help center notes that sustained activity near 90 to 100% for a full 8-hour day can be a sign that something other than a person is generating the input, and suggests opening the Chronicle to see which tools and windows were in use during those periods (Activity KPI).
That kind of check is only fair when employees already know what is monitored and why. Use it as a reason for a conversation, never as a trap.
What to do about it
- Ask before you conclude. Steady input has mundane causes too, such as keeping a laptop awake during a long upload, render or presentation.
- Look at outcomes. If the work is done well and on time, a jiggler says more about the measure than about the person.
- Fix the measure. People reach for jigglers when they are judged on online status or activity percentages. Judge outcomes instead, set idle rules that fit the work, for example a longer idle threshold for meeting and chat apps, and let people explain idle periods. With Intelogos desktop agents, people can be asked whether to keep or discard idle time and add a note such as "In a meeting" (idle time settings).
- Put it in writing. If simulating activity to look busy isn't allowed, say so in your policy, and handle a breach through the same process as any other conduct issue.
What if a remote employee has a second job?
A second job isn't a problem in itself. It becomes one when it creates a conflict of interest, uses company time, equipment or confidential information, or stops someone meeting their agreed hours and results. Handle it with a clear moonlighting policy and a disclosure process, not by searching monitoring data for clues.
Remote work makes it easier to hold two jobs, including two full-time remote jobs at once. Most outside work is harmless, though: weekend shifts, freelancing in an unrelated field, a small business, volunteering. A fair policy separates the two:
- Allow outside work by default, unless it conflicts with the person's duties to you.
- Define a conflict: work for a competitor, customer, supplier or partner in a role that could influence or benefit from their work for you; use of company time, equipment, systems or confidential information; similar work that creates an intellectual property conflict; or outside work that stops them meeting agreed hours, availability or performance expectations.
- Ask for disclosure of outside work that might conflict, before it starts, on a short form.
- Review and decide in writing within a set number of days, with a named person to appeal to.
- Keep disclosures confidential, and don't retaliate against anyone who discloses in good faith.
- Say how monitoring data is used. The clause in our kit includes an option stating that monitoring data isn't used to search for outside employment, and that concerns are raised with the employee directly.
All of this is in the moonlighting and second-job clause in our employee monitoring policy kit, with a disclosure form and notes for HR.
State laws on off-duty conduct vary, so check the rules where the employee works before acting on a second job. New York, for example, bars employers from discriminating against employees for certain lawful activities outside working hours and off the employer's premises, such as political and legal recreational activities, but doesn't protect activity that creates a material conflict of interest with the employer's trade secrets, proprietary information or other business interest (N.Y. Labor Law § 201-d). In California, the Labor Commissioner takes claims for wages lost through a demotion, suspension or discharge for lawful conduct during nonworking hours away from the employer's premises (Cal. Labor Code § 96(k)). Employment contracts and collective agreements can add rules of their own.
This is general information, not legal advice. Laws vary by location and change often; consult your own counsel before acting on it.
How do you monitor remote employees without hurting trust?
Tell people first: what is recorded, what isn't, why, and who sees it. Then use a tool that answers your question with the least intrusive data, let people see their own data, and use it to start conversations rather than to keep score.
Monitoring can backfire when it feels like surveillance. Researchers writing in Harvard Business Review reported that, across two studies, monitored employees were more likely to break rules; they linked this to monitoring weakening people's sense of agency, so they felt less responsible for their own conduct, and employees who felt fairly treated were less affected (HBR, 2022; full paper: Thiel et al., Journal of Management, 2023). For more research on how monitoring is used and received, see our employee monitoring statistics.
Steps to monitor remote workers fairly
- Give notice before you start. Connecticut, Delaware, Maine and New York require employers to notify employees before electronic monitoring, and each state sets its own scope and form of notice (employee monitoring laws by state). Notice is good practice everywhere else too.
- Say why, and what is and isn't collected. Tie monitoring to clear purposes such as balancing workload, preventing burnout or client billing. List what is tracked, what is never tracked and who sees the data.
- Start with team-level patterns. Look at individual detail when a team trend or a specific concern gives you a reason.
- Let people see their own data. It turns monitoring into something they can use, for example to find their most focused hours.
- Use data for support first. Spot overload, rebalance work and recognize strong performers. The first time data is used to punish someone without a conversation, trust is hard to win back.
- Apply it to managers too. When managers are measured the same way, the tool stops feeling like something done to the team.
The free employee monitoring policy kit covers the written notice, the acknowledgment and an FAQ for employees.
This is general information, not legal advice. Laws vary by location and change often; consult your own counsel before acting on it.
How Intelogos shows work patterns without screenshots
Intelogos is a workforce analytics platform that shows how work time is spent across apps and websites, without screenshots or keystroke logging, and lets managers ask questions about it in plain language. Intelogos doesn't take screenshots. Intelogos doesn't record what people type; it records only whether keyboard or mouse input happened.
- What it records. Its agents record the applications and websites used and for how long, window titles, active and idle time and working hours (privacy policy). Window titles and URLs can still contain personal details, such as email subject lines, so say so in your policy.
- Visible or managed agents. Desktop agents run visibly on Windows, macOS and Linux, and people can start and stop them. Background agents are deployed by IT to company computers running Windows or macOS and have no visible interface (understanding agents).
- Defined KPIs. Availability is the portion of the day a person was working, out of the time they were expected to work. Engagement is the portion of tracked time spent in primary and secondary tools. Productivity is the portion of active work minutes spent in primary and secondary tools, out of the time the person is expected to work (understanding performance). Which tools count as work is set for all users and adjusted per department with importance categories.
- Patterns, not proof. Generate Insights reports point out patterns such as after-hours work and sudden changes (Generate Insights), and on the AI Intelligence plan managers can use Ask AI to ask questions such as "Is there anyone who might be at risk of burnout?" (help center).
- Access by role. Managers see the people assigned to them, and the minute-by-minute Chronicle needs a separate permission. Employees can see their own time data, or their full activity and KPIs, depending on an account-wide setting (roles and access).
Plans start at $8 per license per month billed yearly (pricing). See how it works for distributed teams on our remote employee monitoring software page, and read more about employee productivity tracking.
What should you do if you suspect someone isn't working?
Talk to them first, early and in private, with specific examples. Then agree expectations in writing, look at the evidence fairly and document each step before anything becomes formal.
- Check your own setup. Were outcomes and hours agreed in writing? Is the data accurate? Uncategorized tools, idle settings that don't fit the role, or an agent that wasn't running can all make good work look thin.
- Talk first. Raise it as a question, with specifics: "The report due Tuesday slipped, and you've been hard to reach in the afternoons. What's going on?"
- Listen for other causes. Workload, unclear priorities, health, caring responsibilities or problems with tools. Offer support where it fits, and bring in HR where an accommodation may be needed.
- Agree a plan in writing. What changes, by when, what support you'll give and when you'll review it together.
- Review the evidence fairly. Use outcomes, agreed hours and only data employees were told about. Look at several weeks, and apply the same standard you would to anyone else in the role.
- Document each step. Dates, what was discussed, what was agreed and the follow-up. Involve HR before any formal action.
Keep a person in the loop. California's SB 947 (Chapter 859, Statutes of 2026) provides that, from July 1, 2027, an employer may not rely solely on an automated decision system when making a disciplinary or termination decision (California AB 1883, AB 1331 and SB 947 employer guide). A score should start a conversation, never end one.
This is general information, not legal advice. Laws vary by location and change often; consult your own counsel before acting on it.
If the concern is paid time that isn't being worked, our guide to employee time theft includes a sample policy clause and a fair process for that case.
Frequently Asked Questions
Can companies track whether you're working from home?
Yes, on company-owned computers employers commonly use software that records working time, active and idle periods and the applications and websites used; some tools also take screenshots or log keystrokes. The rules depend on the state: Connecticut, Delaware, Maine and New York require employers to notify employees before electronic monitoring (employee monitoring laws by state). Intelogos records work patterns from apps, websites and whether keyboard or mouse input happened; it doesn't take screenshots or record what people type.
This is general information, not legal advice. Laws vary by location and change often; consult your own counsel before acting on it.
How can I monitor employees working from home?
Write a monitoring policy, give employees notice and get their acknowledgment, monitor company-owned devices rather than personal ones, and choose a tool that records only what you need. Then set outcomes for each role, look at team-level patterns first and use individual data to start conversations. The free employee monitoring policy kit has a policy template, a consent form and a state notice checklist, and our remote employee monitoring software page shows how Intelogos supports that approach.
Can my boss tell if I'm working remotely?
If the employer runs monitoring software on the work computer, managers can see what that software records, which varies by tool, and employers should tell employees what it is. In Intelogos, for example, managers see time worked, active and idle time, the applications and websites used and KPIs for the people assigned to them, and the minute-by-minute Chronicle needs a separate permission. Intelogos doesn't take screenshots and doesn't record what people type, and an account-wide setting lets employees see their own data.
Sources
Legal statements were checked on the official pages below on October 23, 2026.
- Connecticut General Statutes § 31-48d: prior written notice to employees of the types of electronic monitoring.
- Delaware Code Title 19, § 705: daily electronic notice, or an acknowledged one-time notice, before monitoring telephone, email or internet use.
- Maine Public Law 2025, chapter 524 (LD 61), 26 M.R.S. § 620-A: notice before employer surveillance begins.
- New York Civil Rights Law § 52-c: written notice on hiring, acknowledged by the employee, and a posted notice.
- New York Labor Law § 201-d: protected off-duty activities and the exception for material conflicts of interest.
- California Labor Code § 96: subsection (k), claims for wages lost through discipline or discharge for lawful off-duty conduct.
- California SB 947 bill text (Chapter 859, Statutes of 2026): no sole reliance on an automated decision system for discipline or termination, from July 1, 2027.
Research, checked September 30, 2026:
- Harvard Business Review: "Monitoring Employees Makes Them More Likely to Break Rules" (Thiel, Bonner, Bush, Welsh and Garud, June 27, 2022).
- Thiel, Bonner, Bush, Welsh and Garud, "Stripped of Agency: The Paradoxical Effect of Employee Monitoring on Deviance," Journal of Management 49(2), 2023: the peer-reviewed paper behind that summary.
Intelogos facts come from our pricing page, privacy policy and help center.