Productivity Calculator

Output per hour = total output ÷ hours worked; minutes per unit = 60 ÷ output per hour; share of target = output per hour ÷ target output per hour × 100; hours at the target rate = total output ÷ target output per hour

Total output

What was produced in the period, in whatever unit the work is counted in: units assembled, calls answered, orders picked, cases closed, dollars of sales. It does not have to be a whole number, because plenty of output is measured in weight, length or money, and the page accepts decimals for exactly that reason. What matters is that the same unit is used for the output, the hours and the target.

Hours worked

The hours that actually went into producing that output, across everyone involved. Lunch, training, waiting for materials and a machine that was down are not hours of production, and counting them as such lowers the figure without anything having gone wrong. When the hours are wrong the productivity figure is wrong in the same direction, and no arithmetic downstream can repair it.

Target output per hour

The rate you are measuring against: a plan, a previous period, a benchmark for the work, or a competitor's figure. It is a field rather than a standard built into the page, because a good rate for one kind of work is a bad one for another — and having it as an input means the same panel answers both what the rate was and how far it is from what was wanted.

Output per hour

The productivity figure itself: output divided by hours. It is the number that can be compared across teams, sites and periods, and the one to look at when a change has been made, because it is the only figure on the panel that is already normalised for how much time went in.

Minutes per unit

The same rate expressed as time rather than quantity, and derived from the printed output per hour rather than from the raw division, so that the two figures on the panel agree with each other. It is the version of the answer that a person on the line can act on: a rate of 30 per hour is a target, and two minutes each is a pace.

Share of target

The achieved rate as a percentage of the target rate. Below 100 means the target was not reached; above it means it was passed, and the figure is not capped there, since beating a target by a factor of three is information rather than an error. A rate low enough next to a high target rounds to zero, which is also printed rather than hidden.

Hours at the target rate

How long the same output would have taken at the target rate. It is the difference between the two states of the world stated in the unit that costs money, which makes it the most useful of the four figures for a decision about whether to invest in reaching the target.

Use it to compare productivity between teams, sites or periods on a scale that is not distorted by how many people or how many hours were involved, and to turn a rate into the pacing a person can work to. It is also the check to run before a staffing decision: the hours figure tells you what is being spent to produce what, and the target column tells you what the same output would cost in time if the rate were met. Three things to keep in mind. Output has to be counted consistently — a change in what counts as a unit will move the figure more than any real change in productivity, and it will look like a performance improvement when it is a definition change. Hours have to be production hours rather than hours present. And the figure is a ratio, so it is silent about quality: a team that produces twice as much per hour by shipping defects has improved its productivity and damaged its business, and nothing on this panel will say so. Note also that two of the four results follow from the same rate, so a change in output or hours moves all four together: there is no combination of inputs where the rate is unchanged and the minutes per unit is not.

1. The default: 1,200 units in 40 hours against a target of 35 an hour

   1. Output per hour: 1,200 ÷ 40 = 30
   2. Minutes per unit: 60 ÷ 30 = 2
   3. Share of target: 30 ÷ 35 × 100 = 85.714 percent, printed as 85.71
   4. Hours at the target rate: 1,200 ÷ 35 = 34.286 hours, printed as 34.29

   The last two lines are the same fact from two sides: the team is at 85.71 percent of the target, and reaching the target would have taken 34.29 hours instead of 40. The 5.71 hours of difference is what the shortfall costs in time, and it is the number to take into a conversation about why the rate is where it is — much more concrete than a percentage.

2. A round number: 900 units in 40 hours against a target of 30

   1. Output per hour: 900 ÷ 40 = 22.5
   2. Minutes per unit: 60 ÷ 22.5 = 2.6667 minutes, printed as 2.67
   3. Share of target: 22.5 ÷ 30 × 100 = 75 percent
   4. Hours at the target rate: 900 ÷ 30 = 30

   A rate of 22.5 per hour is one unit every 2.67 minutes, which is the same statement and a much easier one to check against a clock. This example is also where the rounding is visible: two and two-thirds of a minute is what the work actually takes, and 2.67 is what the panel can print without implying a precision the input does not have.

3. A long period: 2,500 units in 160 hours

   1. Output per hour: 2,500 ÷ 160 = 15.625, printed as 15.63
   2. Minutes per unit: 60 ÷ 15.63 = 3.8388 minutes, printed as 3.84
   3. Share of target: 15.63 ÷ 18 × 100 = 86.833 percent, printed as 86.83
   4. Hours at the target rate: 2,500 ÷ 18 = 138.889 hours, printed as 138.89

   The minutes per unit is worked out from the rounded 15.63 rather than from 15.625, and in this case it makes no difference to the two decimals shown. That is deliberate rather than lucky: the panel is built so that every figure can be checked against the others by hand from the numbers printed, with no last-digit disagreement to explain, and the case where the rounding does show is worth knowing about.

4. The case where the rounding shows: 100.05 units in 30 hours

   1. Output per hour: 100.05 ÷ 30 = 3.335, printed as 3.34
   2. Minutes per unit: 60 ÷ 3.34 = 17.964 minutes, printed as 17.96
   3. Share of target: 3.34 ÷ 3.5 × 100 = 95.428 percent, printed as 95.43
   4. Hours at the target rate: 100.05 ÷ 3.5 = 28.586 hours, printed as 28.59

   This is the example to read if you ever check the panel with a calculator and get a different last digit. Dividing 60 by the unrounded 3.335 gives 17.99 minutes, and the page prints 17.96, because the minutes are derived from the rate as printed rather than from the raw division. The rule behind it is that the panel is internally consistent: 3.34 an hour and 17.96 minutes a unit are the same statement, which they would not be if the two figures were rounded from different starting points.

5. A slow rate against an ambitious target: 100 units in 12 hours

   1. Output per hour: 100 ÷ 12 = 8.333, printed as 8.33
   2. Minutes per unit: 60 ÷ 8.33 = 7.2029 minutes, printed as 7.2
   3. Share of target: 8.33 ÷ 10 × 100 = 83.3 percent
   4. Hours at the target rate: 100 ÷ 10 = 10

   Eight and a third units an hour is one unit every 7.2 minutes, and the target of ten an hour is one every six — a gap of 1.2 minutes per unit that is easier to attack than a percentage. Comparing the first and last lines gives the scale of it too: twelve hours of work that would take ten at the target rate, so the shortfall costs two hours over the period rather than a vague deficit.

A ratio of output to hours is only as meaningful as the two numbers that went into it. Output has to be counted the same way in every period being compared, and any change in what counts as a unit, what counts as finished, or how rework is recorded will move the figure without anything having changed on the floor. Hours have to be hours of production: time waiting, in training or idle because of a breakdown is not time producing, and whether to include it is a decision that has to be made consistently and stated, because it changes the answer. The measure says nothing about quality, safety or cost, so it can improve while the business gets worse — producing more units per hour that have to be reworked is a fall in useful productivity that this arithmetic will record as a rise. It is also a short-run figure: productivity usually falls while a new process or a new team is being learned and rises afterwards, so a single period is a weak basis for a conclusion. And it is not comparable across different kinds of work, which is what the target field is for rather than a table of benchmarks.

Should productivity be measured per person or per hour?

Per hour, whenever the hours are available. Dividing by the number of people makes the figure move for reasons that have nothing to do with productivity: somebody on a part-time week, a new starter in training or a shift that was short-staffed all change it, and the comparison between two sites with different working patterns is meaningless. Hours worked is the denominator the productivity statistics themselves use, and it is the one that answers the question people are actually asking when they ask about worker productivity.

What counts as hours worked?

The hours that went into producing the output, and nothing else: time waiting for materials, in meetings that were not about the work, in training, or idle because equipment was down is not time spent producing. Whether to count indirect time is a real decision with two defensible answers, but it has to be applied the same way every time — a period measured one way and a period measured the other will differ by more than the change being looked for. Whichever convention is used, the arithmetic here cannot tell you which one it was, so write it down next to the figure.

Why are the minutes per unit not what I get from my own division?

Because the minutes are derived from the output per hour as printed, not from the unrounded division. In the example of 100.05 units over 30 hours, the rate is 3.335 and the page prints 3.34; sixty divided by 3.335 is 17.99 minutes, and sixty divided by 3.34 is 17.96, which is the figure shown. The choice is deliberate and it makes the panel self-consistent: the rate and the time per unit on screen are the same statement about the same pace, so a reader who checks one against the other finds them agreeing rather than half a hundredth apart.

Can productivity be compared between two different kinds of work?

Not directly, because the units are different — cases per hour and tonnes per hour are not the same quantity and no arithmetic makes them so. What can be compared is the distance from each area's own target, which is why the target rate is a field on this page: set it to what is expected of that work and the share of target becomes comparable even when the underlying rates are not. That comparison is about how well each area is doing against what was asked of it, which is usually the question behind the request anyway.

Does a higher output per hour mean the business is doing better?

It means more output is coming out of the same hours, which is one component of doing better and not the whole of it. Quality, safety, scrap, rework and the cost of the inputs are all outside this ratio, and a rate can be raised by cutting corners in any of them. Productivity is best read alongside a measure of whether the output was worth producing; on its own it is a rate, and a rate is not a verdict.

Why is the share of target not capped at 100 percent?

Because exceeding a target is information, and how far it was exceeded is the point. A team at 240 percent of a target has either been given a target that was far too low or has changed how the work is done, and both of those are things you want the page to show rather than round down to a satisfied 100. The same reasoning applies at the other end: a rate so small next to its target that the percentage rounds to zero is printed as zero rather than hidden, since that is the case where the target most needs revisiting.

• Principles of Economics 3e, 20.2 Labor Productivity and Economic Growth — defines labor productivity as the value created per unit of input and identifies output per hour worked as the measure of worker productivity — OpenStax, Rice University (United States)
• Principles of Economics 3e, 20.3 Components of Economic Growth — how output per hour, capital per worker and technology combine to raise the output of an economy, which is the same ratio applied at the scale of a country — OpenStax, Rice University (United States)
• Gross Domestic Product — the official measure of total output for an economy, which is the numerator when output per hour is worked out for a country rather than for a team — U.S. Bureau of Economic Analysis (United States)