What it means
The technique is named after an economist who noticed that a minority of the population held the majority of the land, and the same lopsided pattern turns up almost everywhere in business. A few products generate most of the margin, a few customers generate most of the revenue, a few defect types generate most of the complaints, and a few expense categories generate most of the overspend.
Running the analysis takes four steps: choose a single measure such as complaint counts or dollars of cost, group the data into meaningful categories, sort the categories from largest to smallest, and calculate a running cumulative percentage down the list. The point where the cumulative figure passes roughly 80% marks the boundary between the vital few and the useful many.
The reason it earns its keep in finance is that it directs scarce attention. A finance team reviewing 400 general ledger accounts, or a credit controller chasing 900 overdue invoices, cannot treat everything equally, and Pareto ranking reliably shows that a few dozen items carry most of the value at stake.
Two refinements make it far more useful. First, weight by value rather than by count wherever possible, because forty complaints about a $9 accessory matter less than four about a $40,000 machine.
Second, rerun the analysis after acting, since fixing the top causes reshuffles the ranking and a new vital few emerges. The honest limitation is that the split is an observed pattern, not a law.
Real distributions might be 70/30 or 90/10, some problems have no dominant cause at all, and a small category can still be the one that shuts down a production line or loses a licence. Pareto analysis is a way of sequencing work, not a substitute for judgement about severity.
In practice
Real-world examples.
Example
A wholesale distributor ranks its 1,200 customers by gross margin and finds that 84 accounts deliver 79% of total margin. It assigns those accounts named relationship managers and moves the long tail onto a self-service ordering portal.
Example
A finance team investigating a $310,000 budget overspend sorts variances by size rather than by department. Four line items explain $268,000 of the total, so the monthly review is restructured to examine those four in depth and the rest by exception.
Example
A software company categorises 6,500 support tickets from the last quarter. Password resets and one recurring sync failure account for 61% of the volume, which justifies building a self-service reset flow and fixing a single integration bug rather than hiring two more support agents.
Formula
Calculation
Cumulative % = (Running Total of Ranked Category Values / Total of All Categories) x 100, with the vital few being the categories reached before the cumulative figure passes about 80%
A homeware retailer logs 2,000 customer complaints in a month across ten categories. Ranked, the top three are late delivery with 760, damaged packaging with 480 and wrong item shipped with 380, while the remaining seven categories share the other 380 complaints. Check the total: 760 + 480 + 380 + 380 = 2,000.
Cumulative percentages run 760 / 2,000 = 38%, then (760 + 480) / 2,000 = 1,240 / 2,000 = 62%, then (760 + 480 + 380) / 2,000 = 1,620 / 2,000 = 81%. Three of the ten categories account for 81% of complaints. Each complaint costs an average of $45 to handle, so the monthly handling cost is 2,000 x $45 = $90,000, and eliminating 70% of the top three causes would remove 1,620 x 0.70 = 1,134 complaints, worth 1,134 x $45 = $51,030 a month.Case study
Seen in the real world.
Kestrel Machine Tools is a fictional equipment manufacturer used here as an illustrative example. Its warranty costs had climbed to $1,450,000 a year across 38 recorded fault types, and the engineering director's instinct was to launch a quality improvement effort covering all of them.
A Pareto analysis weighted by cost rather than fault count changed the plan completely. Three fault types, a coolant seal failure, a control board firmware defect and a misaligned spindle bearing, accounted for $1,090,000, or roughly 75% of the total, even though they represented only 21% of reported incidents. Two of the three traced back to a single supplier change made eighteen months earlier.
Kestrel reversed the supplier change, issued a firmware update and retrained two assembly cells, at a total cost of about $240,000. Warranty costs in the following year fell to $530,000, and the repeat analysis produced a new top three that had previously been invisible in fourth, fifth and seventh place.
Watch out
Common mistakes.
- Ranking by how often something happens rather than by what it costs, so a high-frequency, low-value issue outranks the small number of expensive ones.
- Treating 80/20 as a precise rule and forcing the data to fit it, when real distributions vary widely and some are genuinely flat.
- Running the analysis once and never repeating it, so the team keeps working on causes that were already fixed months earlier.
Questions
People also ask.
Does Pareto analysis work when there is no dominant cause?
A flat distribution is itself a finding: it tells you the problem is systemic and that picking off individual causes will not help much.
How should categories be chosen?
Make them mutually exclusive and specific enough to act on, because a catch-all category such as "other" that tops the chart tells you nothing useful.
Can it be used for positive things as well as problems?
Yes; ranking customers, products or channels by profit contribution is one of the most common and valuable applications.
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%