Time Tracking Statistics (2026)
Eighty million US workers, 55.6% of everyone on a wage or salary, are paid by the hour, and federal law requires their employer to record the hours they work each day and each week. Federal law does not require a time clock, a particular form, or any software. Among professionals who fill in timesheets, 49% do so daily and 36% weekly or less often, and the weekly fillers are half as likely to be accurate. The Department of Labor recovered $259 million in back wages for 176,957 workers in fiscal 2025, the most since 2019, and 79% of it was unpaid overtime.
The time tracking statistics that rank are a $7.4 billion-a-day figure that is a vendor's 2014 extrapolation from 500 survey responses, a claim that 80% of timesheets are inaccurate that the study never made, and market forecasts presented as measurements. This page uses BLS data on hourly workers, the Fair Labor Standards Act's recordkeeping rules, the one timesheet-habits survey with a stated sample, the BLS research on how people misremember their hours, the court records from a rounding lawsuit, and Department of Labor enforcement data, with sample and date on each. Our time theft statistics page covers the deliberate side; this one covers the honest errors and the rules.
How many workers must have hours recorded
BLS's report on minimum wage workers counted 80.3 million workers aged 16 and over paid at hourly rates in 2024, 55.6% of all wage and salary workers. The 2025 figure from CPS Table 44 is 81.5 million, an eleven-month average excluding October because of the federal shutdown. Of those, 844,000, or 1.0%, were paid at or below the $7.25 federal minimum. On the establishment side, production and nonsupervisory employees were about 82% of private payroll employment in September 2026, a derived ratio from BLS Tables B-1 and B-6. Hourly workers are the population whose hours must be recorded; salaried exempt workers' hours need not be. Our exempt against non-exempt entry covers the line.
What the law requires, and does not
The Fair Labor Standards Act's recordkeeping rule, 29 CFR 516.2, requires employers to keep, for each non-exempt employee, the hours worked each day and the total hours worked each workweek, among other items. The Department of Labor's Fact Sheet 21 states that the Act "requires no particular form for the records" and that "employers may use any timekeeping method they choose." Payroll records must be kept three years and time cards or sheets showing daily start and stop times two years, under 29 CFR 516.5 and 516.6. Rounding is permitted under 29 CFR 785.48 to the nearest five minutes, tenth or quarter hour, "provided that it is used in such a manner that it will not result, over a period of time, in failure to compensate the employees properly." No federal rule requires a time clock, biometric or otherwise. Our guide to calculating hours worked covers the mechanics.
How time is recorded today
The one survey of timesheet habits with a stated sample is AffinityLive's 2014 poll of more than 500 professional services workers in IT, creative, engineering, legal, accounting and public relations, reported by VAR Insights and Harvard Business Review in 2015.
| Measure | Share |
|---|---|
| Fill in timesheets daily | 49% |
| Fill in timesheets weekly or less often | 36% |
| Rarely or never track time spent on email | 51% |
| Track time spent in meetings | 44% |
| Employer uses timesheets to bill clients or set retainers | More than half |
| Accuracy of non-daily fillers relative to daily fillers | Half as likely to be accurate |
It is twelve years old, it covers professional services only, and it is still the only one. Clockify's platform data from 2,300 US companies in 2025 put tracked hours at 66% billable and 34% non-billable, which describes that vendor's users. Our manual against automatic time tracking guide covers the methods.
Accuracy and memory
People remember their hours generously. Frazis and Stewart's 2004 article in the BLS Monthly Labor Review compared self-reported usual weekly hours with time-diary hours and found self-reports about two to three hours higher. It also summarized Mellow and Sider's study matching household reports to employer records, which found workers overreported weekly hours by 3.9% on average and by 11% among managerial and professional workers. Where the diary day fell outside the survey's reference week, self-reports exceeded diary hours by 0.9 to 1.5 hours.
The forgetting-curve figures that vendors quote come from Murre and Dros's 2015 replication of Ebbinghaus, in which one subject memorizing nonsense syllables retained 47% after 20 minutes, 32% after a day and 4% after a month. It is an experiment about syllables, not work, and the AffinityLive finding that weekly fillers are half as accurate is the nearest evidence on timesheets themselves. The "80% of timesheets are inaccurate" line attributed to that study does not appear in it.
Rounding: what the court records show
The best data on what rounding does come from a California lawsuit. In Camp v. Home Depot, decided by the California Court of Appeal in October 2022, a stipulated 10% sample of the company's hourly employees covered 13,387 people, 4.28 million shifts and 516,193 pay periods from 2015 to 2020, with punches captured to the minute and then rounded to the quarter hour. Employees lost minutes on 43.4% of shifts, averaging 3.5 minutes per losing shift, and gained on a similar share, averaging 3.6 minutes. Over pay periods, 47.1% lost time, averaging 10.4 minutes, 49.2% gained, and 3.7% came out exact. The named plaintiff was underpaid by 470 minutes across 1,240 shifts. The court held that where an employer can and does capture exact minutes, quarter-hour rounding that leaves an employee unpaid for time worked is not a defense; it expressly left open rounding where exact capture is impossible. In Donohue v. AMN Services in February 2021, the California Supreme Court barred rounding of meal-period punches and held that records showing non-compliant meal periods raise a rebuttable presumption of violations. Both are California law; federal rounding rules are unchanged. Our timesheet conversion chart covers the arithmetic.
Unbilled time
AffinityLive's 2014 survey estimated that untracked client email alone leaked about $50,000 of revenue per professional per year, that the average professional services firm lost 736 billable hours a year to untracked email and meetings, and, extrapolated across all US professional services workers, 50 million hours or $7.4 billion a day. The last figure is the one that circulates, usually attributed to Harvard Business Review, which only reported it, or presented as measured. It is a vendor's extrapolation from about 500 responses twelve years ago. The underlying habit data in the table above are the usable part. Our billable hours calculator lets a firm estimate its own leakage.
Back wages and penalties
The Department of Labor's Wage and Hour Division publishes its enforcement results each year. In fiscal 2025, per its all-acts data, it recovered $259,294,764 in back wages for 176,957 employees across 16,924 concluded actions, an average of $1,465 per worker and the highest total since 2019, up from $202.7 million in 2024 and $212.3 million in 2023. Civil money penalties rose to $58.7 million from $35.9 million. Under the Fair Labor Standards Act specifically, back wages were $184.4 million, of which $146.4 million was overtime and $27.5 million minimum wage, covering 145,798 employees. Overtime was 79% of FLSA recoveries. Health care, construction and food services led the industry table. Our payroll error statistics page covers the IRS and state figures alongside these.
Adoption and the software market
No primary source measures what share of US employers use time tracking software. The figures that circulate, 43% from Capterra and a 78% accuracy rate from an unnamed ADP survey, could not be traced for this page. Market-size estimates exist but are commercial research with undisclosed methods: Mordor Intelligence put the global time tracking software market at $6.1 billion in 2025 growing to $11.4 billion by 2030, and Grand View Research at $6.1 billion growing to $17.3 billion by 2033 with North America at 38.6% of it. Treat them as vendor forecasts. On features, Clockify's 2025 platform data found 18% of its US business customers using a dedicated time-off feature and 5% a breaks feature. Our tool reviews cover what the software actually does.
Figures to stop using
- "$7.4 billion a day lost to inaccurate time tracking." AffinityLive's 2014 extrapolation from about 500 professional services workers, reported by HBR. Not measured, not economy-wide, and sometimes misquoted as a yearly figure.
- "80% of timesheets are inaccurate." Not in the AffinityLive study. The finding is that non-daily fillers are half as likely to be accurate.
- "75% of businesses experience buddy punching, per the American Payroll Association." No APA source exists; the origin appears to be a 2015 software review site survey.
- "43% of companies use time tracking software, per Gartner." Gartner publishes no such figure.
- "82% of workers have no time management system." A UK training vendor's 500-person panel from 2021.
- "California banned time rounding." Donohue bars rounding for meal periods; Camp applies where exact minutes are captured and left neutral rounding open.
- "DOL recovered $295 million in 2025." The figure is $259.3 million.
- "7% of payroll is lost to time theft and inaccurate tracking." No source.
Key takeaways
- 80.3 million US workers, 55.6% of wage and salary workers, were paid hourly in 2024, and their daily and weekly hours must be recorded under federal law.
- Federal law requires no time clock, no particular form and no software. Rounding to the nearest quarter hour is allowed if it does not systematically underpay.
- 49% of professional services workers fill in timesheets daily and 36% weekly or less, and the weekly fillers are half as likely to be accurate, in the only survey with a stated sample.
- Workers overreport their weekly hours by about 3.9% against employer records, and by 11% among managers and professionals.
- In Home Depot's California records, quarter-hour rounding cost employees minutes on 43.4% of 4.28 million shifts, averaging 3.5 minutes each.
- The Department of Labor recovered $259.3 million in back wages for 176,957 workers in fiscal 2025, the most since 2019. 79% of FLSA recoveries were unpaid overtime.
- The $7.4 billion a day figure is a 2014 vendor extrapolation from about 500 responses. The '80% inaccurate' claim does not appear in that study.
- No primary source measures time tracking software adoption. Market-size figures are commercial forecasts.
Frequently asked questions
What percentage of US workers are paid hourly?
About 56%. 80.3 million workers, or 55.6% of all wage and salary workers, were paid at hourly rates in 2024, per BLS, and 81.5 million in 2025. Hourly workers are the population whose hours must be recorded under the Fair Labor Standards Act.
Does federal law require a time clock?
No. The Fair Labor Standards Act requires records of hours worked each day and total hours each workweek for non-exempt employees, but the Department of Labor states the Act requires no particular form and employers may use any timekeeping method they choose. Time cards must be kept two years and payroll records three.
Is rounding employee time legal?
Federally, yes, to the nearest five minutes, tenth or quarter hour, as long as it does not over time result in underpayment. California bars rounding for meal periods and, where an employer captures exact minutes, has rejected rounding that leaves employees unpaid for time worked.
How much does time rounding cost workers?
In the stipulated sample from the Home Depot lawsuit, covering 13,387 California employees and 4.28 million shifts, workers lost minutes on 43.4% of shifts averaging 3.5 minutes, and gained on a similar share averaging 3.6 minutes. Over pay periods, 47.1% lost an average of 10.4 minutes.
How accurate are timesheets filled in from memory?
Less accurate the longer the gap. Workers who did not complete timesheets daily were half as likely to be accurate as daily fillers in a 2014 survey of more than 500 professionals, and workers overreport their weekly hours by about 3.9% against employer records, per research summarized by BLS.
How much billable time goes unrecorded?
A 2014 survey of more than 500 professional services workers estimated untracked client email leaked about $50,000 per professional per year and 736 billable hours per firm. Its $7.4 billion a day figure is an extrapolation across the US professional workforce, not a measurement.
How much do US employers pay back in wage violations each year?
$259.3 million in back wages to 176,957 workers in fiscal 2025, averaging $1,465 per worker, per the Department of Labor, the highest since 2019. $146.4 million of it was unpaid overtime under the Fair Labor Standards Act, and civil penalties reached $58.7 million.
How long must time cards be kept?
Two years for time cards and sheets showing daily start and stop times, and three years for payroll records, under federal regulations 29 CFR 516.5 and 516.6. Some states require longer.