--- title: "Employee Monitoring Statistics: Surveys and Limits" description: "Compare primary employee monitoring surveys with their dates, countries, samples, and definitions. Separate prevalence, public attitudes, and workplace rules." canonical: "https://idlepilot.com/employee-monitoring-statistics/" updated: "2026-10-28" generated_by: "scripts/build-markdown-twins.mjs" --- > Markdown representation of https://idlepilot.com/employee-monitoring-statistics/ — the HTML page at that URL is canonical and carries the same content. # Employee monitoring statistics In APA’s 2023 U.S. survey, 51% of workers said they were aware their employer used technology to monitor them at work. That measures workers’ awareness, not the share of all employers buying monitoring software. The research below separates reported monitoring, managers’ accounts of workplace tools, and public attitudes, with the population and period kept attached to each result. Sources: [APA: 2023 Work in America survey—AI, monitoring technology, and psychological well-being](#source-apa-monitoring-2023) Last updated: October 28, 2026 Worker and public surveys from the United States, an EU-wide worker survey, an employer survey in six named countries, and ICO-commissioned public research cited in a UK regulator letter. Fieldwork spans 2022–2025. Findings are descriptive; this page does not determine workplace legal rights. ## Findings you can cite | Finding | Value | What it measures | | --- | --- | --- | | U.S. workers aware of employer monitoring | 51% | APA/Harris Poll fieldwork April 17–27, 2023: employed U.S. adults reporting awareness of monitoring by technology, including software, cameras, or scanners. Not an employer census or remote-worker-only survey. Sources: APA: 2023 Work in America survey—AI, monitoring technology, and psychological well-being | | Poor or fair mental health by reported monitoring | 32% versus 24% | In the same April 2023 U.S. worker survey: those aware of monitoring versus those reporting they were not monitored. A cross-sectional association, not evidence that monitoring caused the difference. Sources: APA: 2023 Work in America survey—AI, monitoring technology, and psychological well-being | | Intention to seek a different job | 42% versus 23% | APA’s April 2023 U.S. survey: monitored versus non-monitored workers intending to look for another job in the next year. Intent is not an observed resignation rate. Sources: APA: 2023 Work in America survey—AI, monitoring technology, and psychological well-being | | EU working-hours monitoring | 37% | JRC AIM-WORK survey, 2024–2025 fieldwork, published October 2025: EU workers whose working hours were digitally monitored. This does not mean continuous screen capture or keystroke monitoring. Sources: European Commission JRC: Impact of digitalisation—30% of EU workers use AI | | EU entry and exit monitoring | 36% | JRC AIM-WORK survey, 2024–2025: EU workers whose entry and exit times were monitored. This is a distinct practice from monitoring activity on a computer. Sources: European Commission JRC: Impact of digitalisation—30% of EU workers use AI | | Managers reporting software to monitor working time | 88% in the United States | OECD June–August 2024 survey, published February 6, 2025: managers reporting tools at U.S. firms with at least 20 employees. The U.S. sample was 1,516; the estimate is not the percentage of individual workers monitored. Sources: OECD: Algorithmic management in the workplace—New evidence from an OECD employer survey | | Managers reporting monitoring of communications content and tone | 55% in the United States | Same OECD 2024 fieldwork: managers at U.S. firms with at least 20 employees reporting tools for conversations, voice calls, or emails. A specific tool category, not all forms of monitoring. Sources: OECD: Algorithmic management in the workplace—New evidence from an OECD employer survey | | Public opposition to AI tracking worker movements | 61% | Pew survey of U.S. adults, December 12–18, 2022, published April 20, 2023. Opinion about a described AI use, not respondents reporting their own employer’s practice. Sources: Pew Research Center: Americans’ views on use of AI to monitor and evaluate workers ; Pew Research Center: AI in hiring and evaluating workers—Methodology | | Public opposition to AI tracking desk presence | 56% | Same December 2022 U.S. adult survey: opposition to employers using AI to track when office workers are at their desks. The population includes people not working for pay. Sources: Pew Research Center: Americans’ views on use of AI to monitor and evaluate workers ; Pew Research Center: AI in hiring and evaluating workers—Methodology | | Expected inappropriate surveillance if AI evaluates workers | About eight in ten | Same Pew survey: U.S. adults expecting workers would definitely or probably feel inappropriately watched if AI collected and analyzed information about their work. A hypothetical expectation, not measured prevalence. Sources: Pew Research Center: Americans’ views on use of AI to monitor and evaluate workers ; Pew Research Center: AI in hiring and evaluating workers—Methodology | | Public respondents who would find monitoring intrusive | 70% | ICO-commissioned research cited in its December 21, 2023 letter. Public survey respondents, not employers. The letter does not disclose sample size, fieldwork dates, or geographic sampling frame; its regulatory context is the UK. Sources: ICO letter to the Business and Trade Select Committee, December 21, 2023 | | Public respondents comfortable accepting a monitored job | 19% | Same ICO-commissioned research cited December 21, 2023. Respondents who would feel comfortable taking a new job knowing they would be monitored. Unknown sampling details prevent treating this as a precisely specified national workforce estimate. Sources: ICO letter to the Business and Trade Select Committee, December 21, 2023 | ## How common is employee monitoring? There is no single interchangeable prevalence rate in these studies. A worker can report awareness, a manager can report software available at a firm, and a member of the public can answer a hypothetical question. Each tells you something useful. Pooling their percentages would conceal the differences in who answered and what they were asked. For a statement about worker experience, choose a worker survey and name the practice. For a statement about adoption by firms, choose an employer survey and retain its business-size coverage. For public acceptance, use an opinion survey. The evidence does not support treating those three questions as different routes to the same number. | Evidence | Who answered | What the result describes | | --- | --- | --- | | APA/Harris Poll | Employed U.S. adults | Awareness and reported experiences | | JRC AIM-WORK | Workers across EU Member States | Specific digital monitoring practices | | OECD employer survey | Managers or supervisors at sampled firms | Tools their firms provide | | Pew American Trends Panel | U.S. adults | Opinions about described AI uses | | ICO-commissioned research | Public respondents; sampling details not disclosed in the letter | Attitudes toward monitoring | Sources: [APA: 2023 Work in America survey—AI, monitoring technology, and psychological well-being](#source-apa-monitoring-2023) ; [OECD: Algorithmic management in the workplace—New evidence from an OECD employer survey](#source-oecd-management-2025) ; [Pew Research Center: Americans’ views on use of AI to monitor and evaluate workers](#source-pew-ai-monitoring-2023) ## Basic time records and communications monitoring differ The JRC findings distinguish working-hours monitoring from entry and exit monitoring. Neither label establishes that someone’s screen is being recorded. A useful report names the observed practice rather than describing every timekeeping system as the same kind of surveillance. OECD’s study likewise separates monitoring tasks. Its U.S. estimates concern sampled firms with at least 20 employees, reported by managers. They cannot be converted into the probability that any employee is monitored, and its European comparisons cover France, Germany, Italy, and Spain rather than the whole EU. Larger firms and smaller firms also contribute differently to firm counts and worker counts. Sources: [European Commission JRC: Impact of digitalisation—30% of EU workers use AI](#source-jrc-aimwork-2025) ; [OECD: Algorithmic management in the workplace—New evidence from an OECD employer survey](#source-oecd-management-2025) ## What the wellbeing findings can establish APA’s comparisons show reported experiences occurring together. They do not establish cause. Workplaces can differ in job demands, management, safety requirements, staffing, and other conditions. The survey does not randomly assign people to monitoring. That makes the findings useful for deciding what to investigate, rather than for declaring a universal effect. A practical follow-up would ask what is collected, what employees understand about it, and how managers use it. When reporting the job-search result, preserve the distinction between intending to look for work and actually leaving an employer. Sources: [APA: 2023 Work in America survey—AI, monitoring technology, and psychological well-being](#source-apa-monitoring-2023) ## Public opinion does not measure software adoption Pew asked adults about described uses of AI, including workplace movement and desk tracking. These are questions about acceptance and expected consequences. An answer can be meaningful even when the respondent has never encountered that technology. It would therefore be misleading to say the opposition figures reveal how many employers use it. ICO’s published letter supplies another public-attitude perspective, but provides fewer survey details. The figures are included with that gap visible. They should not be used to rank the UK against the United States: the question wording, populations, and sampling information are not aligned. Uncertainty about methods should travel with a statistic when it is quoted. Sources: [Pew Research Center: Americans’ views on use of AI to monitor and evaluate workers](#source-pew-ai-monitoring-2023) ; [ICO letter to the Business and Trade Select Committee, December 21, 2023](#source-ico-letter-2023) ## Workplace rules are a separate question A survey finding that a practice occurs does not establish that it is permitted in a particular workplace. A finding that people dislike a practice does not establish a legal prohibition. The ICO letter discusses regulatory guidance separately from its commissioned research; those are different forms of evidence. This page reports survey results, not an assessment of an individual employer’s compliance. If you need to understand your own situation, identify the country, relevant policy, monitoring purpose, device, and data involved. A company’s technology documentation or a broad monitoring percentage cannot, by itself, settle that question. Sources: [ICO letter to the Business and Trade Select Committee, December 21, 2023](#source-ico-letter-2023) ## How to cite monitoring evidence without overstating it Keep the respondent in the sentence. Write that managers reported a tool at their firms, or that workers reported being aware of monitoring. Include the fieldwork period as well as the publication date when they differ. This is especially useful for the OECD and Pew studies, which were published after their surveys took place. Keep the monitoring category equally visible. Tracking attendance, analyzing communications, observing location, and evaluating performance can involve different systems and purposes. A statistic about one category is not evidence for every other category. This approach gives readers a clear claim they can check and prevents a dramatic percentage from becoming broader than the underlying research. Sources: [OECD: Algorithmic management in the workplace—New evidence from an OECD employer survey](#source-oecd-management-2025) ; [Pew Research Center: AI in hiring and evaluating workers—Methodology](#source-pew-ai-method-2023) ; [European Commission JRC: Impact of digitalisation—30% of EU workers use AI](#source-jrc-aimwork-2025) ## Method Read original publisher findings and available methods: APA/Harris Poll, Pew’s American Trends Panel, the European Commission JRC, OECD’s employer-survey report, and the ICO’s own dated letter. Record fieldwork and publication separately. Preserve employee versus public versus manager populations and disclose missing sample information instead of supplying an assumed denominator. Use category-specific findings. Do not average surveys, infer a world prevalence rate, or use legislation as evidence that a practice is common. ### Limitations - All selected surveys rely on respondents’ knowledge or opinions. They are not independent device audits. - APA’s cross-sectional associations do not prove causation. Subgroup sample sizes are not disclosed in the cited findings page. - OECD covers firms with at least 20 employees in six countries. Its Europe grouping is not EU-wide. Basic software need not use AI. - Pew surveyed the U.S. adult public, including nonworkers. The ICO letter omits key survey-method details. These sources cannot establish what a particular employer does or what a particular workplace rule permits. ## What we could not establish - A universal claim that 80% or 90% of all employers monitor every employee is excluded without a matching representative source and definition. - Claims that remote work caused all monitoring growth, or that monitoring necessarily causes burnout or resignations, are excluded. - A single global bossware adoption rate, and claims that presence indicators reveal all monitoring taking place, are excluded. - Rules about notice, consent, or legality in an unspecified jurisdiction are not presented as statistics or legal advice. ## FAQ What percentage of workers are monitored? It depends on the population and definition. APA measured U.S. workers’ awareness of technology monitoring; JRC measured particular digital practices among EU workers. Neither supplies a universal workforce rate. Does the OECD result mean almost all U.S. employees are watched? Its respondents were managers reporting tools at firms with at least 20 employees. A firm-level adoption estimate is not a count of employees exposed, and tools cover different managerial tasks. Is recording working hours the same as recording a screen? The studies distinguish time monitoring from other practices. A timekeeping statistic does not establish screen capture, communications analysis, or keystroke collection. Do the surveys prove monitoring harms mental health? APA reports an association, not a confirmed cause. Other differences between workplaces and workers could contribute to the pattern. Are Pew’s percentages based only on employees? No. The cited study surveyed U.S. adults. Its headline monitoring figures concern opinions and expectations about AI, rather than an audit of employer adoption. Do monitoring statistics show whether my employer can monitor me? No. Descriptive research and workplace rules answer different questions. The ICO’s letter presents its public research alongside a separate discussion of regulatory guidance. ## Sources and citation notes Cite the original source and its date for each figure. For our own studies, include the sample and limitations above. 1. [APA: 2023 Work in America survey—AI, monitoring technology, and psychological well-being](https://www.apa.org/pubs/reports/work-in-america/2023-work-america-ai-monitoring) Published: 2023-09; exact day not stated Population or sample: 2,515 employed U.S. adults surveyed online by The Harris Poll for APA, April 17–27, 2023; opt-in respondents, weighted. Monitoring subgroup counts not disclosed on this page. Scope: United States; self-reported monitoring awareness, mental health, and job-search intentions. Broad technology definition includes cameras and scanners. Checked: 2026-10-28 2. [European Commission JRC: Impact of digitalisation—30% of EU workers use AI](https://joint-research-centre.ec.europa.eu/jrc-news-and-updates/impact-digitalisation-30-eu-workers-use-ai-2025-10-21_en) Published: 2025-10-21 Population or sample: AIM-WORK survey: 70,316 workers across all 27 EU Member States, conducted in 2024–2025. Exact item-response denominators are not provided in the news summary. Scope: EU-wide worker survey; selected findings concern working-hours and entry/exit monitoring. Underlying report record JRC143072 dated October 13, 2025. Checked: 2026-10-28 3. [OECD: Algorithmic management in the workplace—New evidence from an OECD employer survey](https://www.oecd.org/content/dam/oecd/en/publications/reports/2025/02/algorithmic-management-in-the-workplace_3c84ed6d/287c13c4-en.pdf) Published: 2025-02-06 Population or sample: 6,047 firms with at least 20 employees, interviewed June–August 2024 through managers/supervisors. Country samples: U.S. 1,516; France 1,006; Germany 1,010; Italy 1,009; Spain 1,006; Japan 500. Weighted survey; Annex A. Scope: Six named countries, all sectors. Firm-level tool availability reported by managers; European grouping covers four countries, not the whole EU. Working-time and communications categories: report page 25. Checked: 2026-10-28 4. [Pew Research Center: Americans’ views on use of AI to monitor and evaluate workers](https://www.pewresearch.org/internet/2023/04/20/americans-views-on-use-of-ai-to-monitor-and-evaluate-workers/) Published: 2023-04-20 Population or sample: 11,004 U.S. adults in the American Trends Panel, surveyed December 12–18, 2022; weighted probability-based panel. Scope: United States, adults aged 18 and older; public opinions and hypothetical expectations, not employer adoption. Checked: 2026-10-28 5. [Pew Research Center: AI in hiring and evaluating workers—Methodology](https://www.pewresearch.org/internet/2023/04/20/ai-in-hiring-methodology/) Published: 2023-04-20 Population or sample: 11,004 respondents from 12,448 sampled panelists; full-sample margin of sampling error ±1.4 percentage points. Scope: Noninstitutionalized U.S. adults aged 18 and older, including Alaska and Hawaii; December 12–18, 2022 fieldwork. Checked: 2026-10-28 6. [ICO letter to the Business and Trade Select Committee, December 21, 2023](https://ico.org.uk/media2/b3upftrl/ic-380283-q2q7-attachment.pdf) Published: 2023-12-21 Population or sample: Not disclosed: sample size, respondent eligibility, sampling method, geographic sampling frame, and fieldwork dates are absent from this letter. It reports ICO-commissioned public research. Scope: UK regulator context; public attitudes reported in the Employee Monitoring section on page 2. Do not assume the sample comprises current employees or that it is nationally representative. Checked: 2026-10-28 ## Related guides and tools - [What workplace tools document collecting](https://idlepilot.com/research/can-employers-detect-mouse-jigglers/) - [Remote work statistics](https://idlepilot.com/remote-work-statistics/) - [Microsoft Teams statistics](https://idlepilot.com/microsoft-teams-statistics/) - [Slack statistics](https://idlepilot.com/slack-statistics/) - [Teams presence and availability](https://idlepilot.com/keep-teams-active/) ## Keep Slack or Teams Available during saved hours, even with your computer off. [Try Idle Pilot free](https://app.idlepilot.com) 3-day free trial · No credit card