What it means
An average on its own hides the shape of the data. A fund averaging 6% a year by returning between 4% and 8% is a very different proposition from one averaging 6% by alternating between -20% and 32%, and only a measure of spread reveals the difference.
The common measures are range, which is simply the highest value minus the lowest, variance, which is the average of the squared distances from the mean, and standard deviation, which is the square root of variance and is expressed in the same units as the original data. Standard deviation is the one quoted most often because it is directly comparable to the average itself.
In finance, dispersion of returns is the standard proxy for risk, and it feeds into ratios that compare reward against variability. In operations it is often more useful than the average, because customers experience the worst week rather than the mean week, and in sales it exposes whether results come from a repeatable process or from two exceptional individuals.
A related figure is the coefficient of variation, which divides the standard deviation by the mean to give a relative measure. That matters when comparing things of different sizes, since a standard deviation of $50,000 means something very different on a $200,000 base than on a $20,000,000 one.
The main caution is that dispersion treats upside and downside identically. A manager who beats expectations wildly in good years is penalised by the same measure as one who collapses in bad years, which is why downside-only measures are sometimes preferred.
In practice
Real-world examples.
Example
A software company reviews quota attainment across 30 sales representatives and finds an average of 98% with a standard deviation of 41%. The high dispersion shows the number is carried by five outperformers while eleven people sit below 60%.
Example
A parcel network reports an average delivery time of 2.1 days, but the standard deviation of 1.4 days means a meaningful share of parcels arrive on day four or five. Customer complaints track the spread, not the average.
Example
An investment committee compares two funds that both returned 7% a year over a decade. The first has a standard deviation of 8% and the second 19%, so the committee allocates to the first because its pension liabilities cannot absorb the swings.
Formula
Calculation
Sample standard deviation = the square root of [ the sum of (each value - the mean) squared, divided by (n - 1) ]
A portfolio produces annual returns of 4%, 12%, -6%, 18% and 2% over five years.
Mean: (4 + 12 - 6 + 18 + 2) / 5 = 30 / 5 = 6%
Deviations from the mean: -2, 6, -12, 12, -4
Squared deviations: 4, 36, 144, 144, 16
Sum of squared deviations: 4 + 36 + 144 + 144 + 16 = 344
Sample variance: 344 / (5 - 1) = 344 / 4 = 86
Standard deviation: the square root of 86 = about 9.3%
So the average is 6% but a typical year lands roughly 9.3 percentage points either side of it. The range tells a similar story: 18% - (-6%) = 24 percentage points between best and worst.Case study
Seen in the real world.
Blue Harrow Logistics is an invented business used for this illustrative case. Its board celebrated an average warehouse pick accuracy of 99.1% across six depots until an analyst broke the figure down and found individual depot results of 99.8%, 99.7%, 99.6%, 99.5%, 99.4% and 96.6%.
The mean of those six figures is 594.6 divided by 6, which is 99.1%, exactly as reported. But five depots clustered within 0.3 percentage points of each other while one sat almost three points below, and that single depot generated the majority of the group's customer credits.
Once the board looked at dispersion rather than the average, the fix became obvious and cheap: retrain and re-equip one site rather than launch a group-wide quality initiative. In this illustrative example the average had been actively misleading, because it described a depot that did not exist.
Watch out
Common mistakes.
- Reporting an average without any measure of spread, which hides whether the result is typical or the midpoint of two extremes.
- Comparing standard deviations across data sets of very different size instead of using the coefficient of variation.
- Assuming low historical dispersion guarantees low future risk, when quiet periods often precede sharp moves.
Questions
People also ask.
What is the difference between variance and standard deviation?
Variance is the average squared distance from the mean, and standard deviation is its square root, which returns the figure to the same units as the original data.
Should I use the sample or population formula?
Use the sample version, dividing by n minus 1, when your data is a sample of a larger set, which covers almost all business situations.
Does high dispersion always mean high risk?
Not necessarily, because the measure counts unusually good outcomes as well as bad ones, which is why downside deviation is sometimes used instead.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%