What it means
The measure is built from the Lorenz curve, a chart that plots the cumulative share of a population against the cumulative share of whatever is being measured. If everyone had an identical share, the curve would be a straight diagonal line.
The more the actual curve sags below that diagonal, the more unequal the distribution. The coefficient is simply the area of that sag expressed as a proportion of the whole area beneath the diagonal.
Because it is a ratio, it strips out the size of the population and the size of the total, which makes it directly comparable between a country of 80 million people and a customer book of 400 accounts. In business, the most useful application is concentration risk.
Working out the Gini coefficient of revenue across customers tells you in one figure whether losing a handful of accounts would be an inconvenience or a crisis, and tracking it over time shows whether a sales strategy is broadening the base or narrowing it. Credit teams use a closely related version to judge how well a scoring model separates good borrowers from bad ones.
A model with a Gini near zero is no better than a coin toss, while values in the range often seen for well-built consumer scorecards indicate the model is ranking risk usefully. The main limitation is that very different distributions can produce the same score, because a single number cannot describe the shape of a curve.
A country where the top 1% dominates and one where the bottom half is unusually poor can score alike, so the coefficient should be read alongside the underlying shares rather than instead of them.
In practice
Real-world examples.
Example
A software company calculates a customer revenue Gini of 0.71 and discovers that its top 20 accounts out of 900 drive most of its income. The board sets a target of reducing the figure to 0.60 within two years by building a lower-priced tier for smaller firms.
Example
A bank's risk team compares two credit scorecards on the same portfolio and finds Gini values of 0.34 and 0.51. The second model is adopted because it ranks defaulters far more effectively at the same approval rate.
Example
A national retailer measures the Gini coefficient of sales across its 220 stores and finds it has crept up over three years. The pattern points to a small group of flagship sites carrying the estate while the long tail stagnates.
Think of it
“Gini coefficient measures inequality-0 is perfectly equal, 1 is perfectly unequal.
Formula
Calculation
Gini = area between the line of equality and the Lorenz curve, divided by the total area beneath the line of equality. For grouped data it can be computed as Gini = 1 - the sum over each group of (change in cumulative population share) x (sum of the current and previous cumulative value shares).
A business splits its 500 customers into five equal groups of 100, ranked by annual spend. The groups account for 4%, 8%, 16%, 24% and 48% of revenue, so cumulative revenue shares are 0.04, 0.12, 0.28, 0.52 and 1.00, while each group adds 0.20 to the cumulative customer share.
The five terms are 0.20 x (0.04 + 0) = 0.008, then 0.20 x (0.12 + 0.04) = 0.032, then 0.20 x (0.28 + 0.12) = 0.080, then 0.20 x (0.52 + 0.28) = 0.160, and finally 0.20 x (1.00 + 0.52) = 0.304. These sum to 0.584, so the Gini coefficient is 1 - 0.584 = 0.416, or about 0.42, indicating meaningful but not extreme concentration.Case study
Seen in the real world.
The following is an illustrative and fictional example. Ashbourne Logistics, an invented freight brokerage, had grown revenue by 40% over three years and its leadership assumed the business was becoming steadily safer. A new finance director calculated the Gini coefficient of revenue across the client base and found it had risen from 0.48 to 0.76 over the same period.
The fictional analysis showed that almost all the growth had come from two large retail contracts, while the number of mid-sized clients had actually fallen. Losing either contract would have removed more gross profit than the entire remaining client base generated.
Ashbourne's illustrative response was to price a smaller-shipment service aimed at regional wholesalers and to report the concentration figure at every board meeting alongside revenue, so that growth and fragility were never again read as the same signal.
Watch out
Common mistakes.
- Reading a higher Gini coefficient as automatically bad, when in some contexts, such as a credit scoring model, a higher figure means better discrimination.
- Comparing coefficients calculated on different groupings, since splitting data into five bands rather than a hundred changes the result.
- Relying on the single number without looking at the underlying shares, because very different distributions can produce identical scores.
Questions
People also ask.
What counts as a high Gini coefficient for customer revenue?
There is no universal threshold, but figures above roughly 0.70 usually signal that a small number of accounts carry the business.
Can the coefficient be negative?
Not in the standard form, since it is bounded between 0 for perfect equality and 1 for total concentration.
Is it the same as the 80/20 rule?
They describe related ideas, but the Gini coefficient uses the whole distribution while the 80/20 rule quotes a single point on the curve.
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%