Productivity has an exact meaning in official statistics: how much output is produced for each hour of work. This article explains that definition, walks through the latest US numbers, and shows how a team can use the same logic in everyday work.
What does “productivity” actually mean?
In everyday talk, being productive means being busy, fast or always available. Statistical agencies use a narrower idea. The US Bureau of Labor Statistics (BLS) says its productivity office “measures how efficiently the U.S. converts inputs into the outputs of goods and services.” Labor productivity, the best-known measure, is output per hour worked: it compares how fast output grows with how fast hours worked grow.
The definition has two parts, and both matter. Output is what gets produced, such as goods sold, services delivered or value added. Hours are the labor put in to produce it. Productivity rises when output grows faster than hours, and it falls when hours grow faster than output. The UK Office for National Statistics describes labour productivity in similar terms, as the efficiency of the workforce, and publishes it per worker, per job and per hour.
Two things follow from this. First, working longer does not by itself raise productivity, because extra hours enlarge the bottom of the fraction. Second, a team can become more productive while working the same hours, or the same number of hours less, if each hour produces more useful results.
What do the latest figures show?
As of its most recent release, the BLS reported results for the second quarter of 2026 for the US nonfarm business sector (essentially private-sector business, excluding farms). The figures below come from the agency’s own productivity page.
| Measure, nonfarm business, Q2 2026 | Change reported by BLS |
|---|---|
| Real value-added output | +1.7% |
| Hours worked | +0.3% |
| Labor productivity (output per hour) | +1.4% |
| Hourly compensation (revised) | +2.6% |
| Unit labor costs (revised) | +1.2% |
The BLS also published productivity figures for specific industries for 2025. According to the agency, labor productivity rose in 42 states and the District of Columbia, in wholesale trade by 4.4% and in retail trade by 2.9%. In manufacturing and mining, productivity rose in 39 of 80 manufacturing industries and 2 of 5 mining industries. Among selected service-providing industries, it rose in 15 of 30.
That last set of numbers is a useful reminder. Averages hide a lot of variation. In the same year, roughly half of the measured manufacturing and service industries saw gains while the rest did not. A headline number says little about any one team.
The next BLS release, covering the third quarter of 2026 on a preliminary basis, is scheduled for November 5, 2026.
How do the numbers fit together?
The table is internally consistent, and the links between rows are worth understanding because they apply to any team.
Output, hours and productivity
If output grows 1.7% and hours grow 0.3%, productivity grows by roughly the difference. Illustrative arithmetic: 1.017 divided by 1.003 is about 1.014, which is a gain of about 1.4%, matching the reported figure. Nearly all the extra output came from more output per hour, not from more hours.
Pay, productivity and cost
Unit labor costs measure how much labor cost goes into each unit of output. When hourly compensation grows faster than productivity, unit labor costs rise. Illustrative arithmetic: 1.026 divided by 1.014 is about 1.012, a rise of about 1.2%, which again matches the reported figure. Pay grew 2.6% per hour, productivity grew 1.4%, and the gap shows up as higher cost per unit.
This relationship explains why productivity is not only an efficiency topic. It connects to wages, prices and a business’s margins, which is why governments and central banks watch it closely.
Why do hours and output not always move together?
Anyone who has worked a very long day knows the last hours are rarely the best ones. Fatigue, errors and rework mean that the tenth hour of a day usually produces less than the third. Output per hour is therefore not a constant, and pushing hours up can drag the average down.
Several other factors move the ratio:
- Tools and methods. Better software, templates or processes let the same hour produce more.
- Skills and training. People who know a task well finish it faster and with fewer mistakes.
- Organization of work. Clear priorities and fewer handoffs cut waste, a point covered in a simple system for managing tasks.
- Interruptions. Constant context switching lowers the quality of each hour, as discussed in deep work vs busy work.
- Meetings. Time spent in gatherings with no clear outcome counts as hours worked but adds little to output, which is why running meetings that respect time matters.
None of these requires working harder. They all change what each hour yields.
Can a small team measure its own output per hour?
A team can borrow the logic of the statistic, but it needs care. National output is measured in money, adjusted for price changes. A team usually has to choose its own unit of output, and the choice shapes behavior.
Pick an output that reflects value
Good candidates are things a customer or colleague would recognize: support tickets resolved, reports delivered, features shipped, invoices processed. Poor candidates are things that are easy to count but easy to inflate, such as emails sent or lines of code. If the unit can be gamed, it will be.
Track hours honestly
Use the hours the team actually spends, including meetings and rework, not the hours on the schedule. Weekly totals are enough. Daily tracking tends to feel like surveillance and distorts the numbers.
Look at the trend, not the single week
One week proves nothing. A holiday, a hard bug or a sick colleague will move the ratio. Compare month against month, and ask what changed when the figure moves.
What should a team do with the answer?
A worked example shows the idea. Imagine a four-person support team that resolves 480 tickets in a 160-hour week of combined effort. That is 3 tickets per hour. After the team agrees on shorter written replies to common questions and cuts a weekly status meeting, it resolves 510 tickets in 150 hours, or 3.4 tickets per hour. Output rose 6.25% while hours fell 6.25%, and output per hour rose about 13%. The numbers are invented for illustration, but the pattern is realistic: removing low-value hours can help as much as adding effort.
The temptation after seeing a rising number is to raise targets immediately. That tends to backfire. If a team is rewarded for gains with a bigger quota, people learn to hide improvements. A better approach is to decide in advance what a gain buys: faster response times, fewer late evenings, time for training, or a sooner finish to a project.
Where does this measure fall short?
Output per hour is a useful lens, but it has limits that any manager should keep in view.
- Quality is hard to count. A faster process that produces worse results looks productive on paper.
- Creative and long-horizon work resists measurement. Research, strategy and mentoring pay off over months, not weeks.
- Averages mislead. The BLS industry results above show that gains are uneven even in the same year.
- It is not a measure of worth. The statistic describes a process, not the value of a person.
Remote and hybrid arrangements add another wrinkle, since hours worked can be harder to observe. For how people actually spend their time, see what time-use data says about working from home. The honest conclusion from that research and from the statistics here is the same: measure the work, not the presence.
A lens for choosing which hours to improve
Productivity, defined as output per hour, gives teams a precise question to ask: where are our hours going, and what do they produce? The latest BLS figures show a business sector where output rose faster than hours in the second quarter of 2026, but also one where pay rose faster than productivity, which is a reminder that efficiency, wages and costs are linked. For a team, the practical move is modest. Pick an output that reflects real value, count the hours honestly, watch the trend, and cut the hours that produce little before asking anyone to work more of them.
This article is general information about work and productivity measurement, not business, financial or legal advice.
Sources: US Bureau of Labor Statistics; UK Office for National Statistics
