Around 80% of a startup's results tend to come from around 20% of its inputs.
The rule's proper name is the Pareto principle. It describes a pattern rather than a rule of nature: effort and reward are rarely spread evenly, so a minority of customers, channels or tasks tends to carry most of the result. For founders it works best as a measuring habit, not a productivity slogan.
The ratio itself is the least interesting part. What matters is the shape underneath it, where that shape turns up in a young company, and the two or three places where applying it will quietly damage your business.
What is the 80/20 rule actually called?
The 80/20 rule is called the Pareto principle.
It takes its name from Vilfredo Pareto, an Italian economist who showed in his Cours d'économie politique (1896 to 1897) that roughly 80% of Italy's land was held by about 20% of the population. He looked at other countries and found a similar pattern.
Pareto was not writing about startups, marketing or calendars. He was describing how wealth distributes itself across a population.
You will also see the same idea written as the 80 20 principle, the Pareto rule, or the principle of factor sparsity. Operations and software teams often shorten it to Pareto 80/20.
So what is the Pareto principle, in one sentence? It is the observation that outcomes are distributed unevenly, so a minority of causes carries most of the effect.
The man who named it admitted he got the name wrong
Here is the part almost nobody mentions.
Pareto did not name the Pareto principle. Joseph M. Juran did, in the late 1940s, while preparing his Quality Control Handbook. Juran was a quality engineer who had spotted that a small share of causes produced most manufacturing defects.
He needed a short label for what he had found, and he borrowed Pareto's name for it. Decades later he owned up to the error.
In a 1974 paper called The Non-Pareto Principle: Mea Culpa, Juran wrote that he "was forced to confess that I had mistakenly applied the wrong name to the principle". Had he been "structured along different lines", he said, he "would have called it the Juran principle".
So why should a founder care about a naming dispute from 1974? Because of the second thing Juran changed.
His original phrase for the pattern was "the vital few and the trivial many". He later rewrote it as "the vital few and the useful many", because describing the remaining 80% as trivial was wrong and he did not want readers to treat it that way.
That one word is the whole argument of this article. The 80% is not waste. It is simply not where your returns are concentrated, which is a different claim entirely.
Why 80 and 20 do not have to add up to 100
Plenty of people assume the two numbers have to sum to 100. They do not.
They are measuring different things. One is a share of outputs, the other a share of inputs. That they happen to total 100 is a coincidence, and probably the reason the shorthand stuck so well.
Real distributions wander. You might find 70/20, 90/10, or in extreme cases 99/1. One enterprise account can out-earn the other forty put together, and nothing about that breaks the principle.
Sitting underneath the shorthand is a long-tailed statistical distribution, the Pareto distribution. The particular version that produces a neat 80/20 split carries a Pareto index of about 1.16. Shift the index and the split shifts with it.
The practical consequence is this. Do not test whether your business obeys 80/20. Measure how steep your curve actually is. A founder who knows that four customers out of fifty produce 71% of revenue is holding something far more useful than a slogan.
Pareto principle examples: where the 80/20 rule shows up in a business
Concentration tends to appear anywhere you can rank things and add them up.
| What you measure | What the concentration usually looks like |
|---|---|
| Revenue | A handful of customers carry most of it |
| Churn | A few reasons explain most cancellations |
| Support | A small set of issues generates most tickets |
| Product usage | Most sessions touch a thin slice of the features |
| Marketing | One or two channels bring most qualified signups |
| Sales pipeline | A minority of deals holds most of the closed value |
| Engineering | A few parts of the codebase absorb most of the work |
| Bugs | A small number of defects cause most reported failures |
Treat that table as a list of things to check, not a list of facts. The pattern is common, not guaranteed. Some startups have genuinely flat revenue curves, and finding that out is itself worth knowing.
How to find your own 20% in an afternoon
This is where most articles stop and most founders get stuck. The method is short enough to run between meetings.
- Pick one output you care about. Revenue this quarter, support hours, closed deals, active users. One at a time.
- List every input that produced it. Every customer, channel, feature or task. Use real records, not memory.
- Rank them from largest to smallest. A spreadsheet column is enough.
- Add a running total as a percentage of the whole. Cumulative share is the number that reveals the shape.
- Find the row where the running total passes 80%. Count how many inputs sat above it, and work out what share of your list that is.
- Write down the ratio you actually found. Not 80/20. Yours.
Here is what that looks like for a small company with ten paying customers and £100,000 of annual revenue.
The figures above are illustrative, chosen to show the arithmetic cleanly. Run the same exercise on your own numbers and the curve will have its own slope.
Once you can see the curve, the decisions get easier. Customers one and two deserve a named owner, a quarterly review and a plan for what happens if either leaves. Customers seven to ten probably do not deserve a bespoke feature.
The same method works on the input side of a startup too, which is why it pairs well with the ordinary work of finding one person who will pay before building anything bigger.
Pareto analysis and the Pareto chart
The exercise above has a formal name: Pareto analysis.
It came out of quality management rather than startup advice, which is why Juran kept returning to it. A factory trying to cut defects would rank the causes, find the few producing most failures, and fix those first.
The standard way to draw it is a Pareto chart. Ranked bars show each input from largest to smallest, and a line across the top tracks the running total as a percentage. Where that line crosses 80% is your answer.
The chart above is the bar half of one. Add the cumulative line and you have the whole picture in a single image, which is usually the quickest way to show a co-founder or a board where the business actually earns its money.
Pareto analysis is deliberately blunt. It tells you where the mass sits, not why it sits there. A customer may be large because of one contract that ends in March, and no ranking will warn you about that. Treat the chart as the start of the question rather than the answer.
What concentration looks like in UK startup funding
Founders meet the Pareto pattern from the other side when they raise money, because venture returns are concentrated too.
The British Business Bank's research on early-stage venture capital returns reported a pooled distributed-to-paid-in ratio of 1.43 for early-stage UK funds, against a pooled total value to paid-in of 1.99. More tellingly, funds based outside London, Cambridge and Oxford showed a pooled DPI of 1.65 compared with 0.85 inside that triangle, a gap the Bank attributes to a small number of funds performing exceptionally well.
That is the shape in miniature. A pooled average hides the fact that a few funds, and inside them a few companies, carry the result.
It explains investor behaviour that otherwise looks irrational. An investor who needs one company in a portfolio to return the entire fund will push a founder towards the larger market, even when the smaller one looks safer. They are not being greedy. They are pricing a very long tail.
It also explains why the survival numbers are so unforgiving. UK five-year business survival data from the Office for National Statistics shows how few businesses are still trading after five years, and we have covered why most startups fail in more detail elsewhere.
When the 80/20 rule does not work for startups
This is the section the rest of the internet leaves out, and it is the one that will save you money.
Before product-market fit, you do not have the data. The 80/20 rule is a method for analysing results you already have. A startup with eleven customers and three months of history has nothing to rank. Applying the rule here means guessing which 20% matters and then defending the guess, which is worse than admitting you do not know. At that stage the honest move is validating your startup idea before you write code, not optimising a curve that does not exist yet.
Your current 20% was once in the 80%. Today's best channel was a side experiment eighteen months ago. Cut every low-performing input and you also cut the pipeline that produces next year's winners. A reasonable split is to protect most of your effort for the known few and reserve a visible slice for deliberate experiments, with the understanding that most of them will fail.
Concentration is a risk report, not a trophy. If two customers produce 80% of revenue, that is not an achievement. It is a dependency. One leaves and your company has a different problem than the one it had yesterday. The right response is often the opposite of doubling down: look after those accounts, then work hard on the next ten.
Survivor data distorts the ranking. You can only rank the customers who stayed, the channels you tried and the features you shipped. The best channel you never tested does not appear. That blind spot is the same one we described in survivorship bias in startup advice, and the Pareto method is quietly vulnerable to it.
The 4% hiding inside the 80/20 rule
Apply the rule to its own output and the arithmetic gets uncomfortable.
If 20% of inputs produce 80% of results, then running the same logic on that top slice suggests 4% of inputs produce 64% of results. Square it again and about 0.8% of inputs produce roughly half.
This is arithmetic, not a measured finding, and real curves rarely behave so politely. But it holds a real lesson: the top of your list is almost certainly more concentrated than the top-level ratio implies.
In practice that means your single best customer, your single best channel or your single best engineer may matter more than the whole bottom half of the list combined. Most founders spend their attention in roughly the opposite proportion.
The 80% you cannot delete
Juran's correction matters most here.
Some of the lowest-return work in a startup cannot be cut at any price. Payroll, VAT returns, Companies House filings, security patches, contracts, backups. None of it appears in the 20% that drives growth. All of it will end your company if ignored.
The same is true of the parts of a product that nobody praises. Password resets and invoices rarely delight anyone, and their absence loses customers.
So the practical question is not "what can I delete?" It is sharper than that:
- What belongs in the few, and deserves my own time?
- What is necessary but low-return, and should be automated, batched or handed over?
- What is genuinely optional, and can stop today?
Only the third group gets cut. The second group is where most of the time savings actually come from, and it overlaps heavily with the ordinary ways to save time in an early-stage startup that founders put off for too long.
A weekly habit that keeps it honest
The principle stops working the moment it becomes a one-off audit, because your curve keeps moving.
A short Friday routine keeps it current. Look at the week's closed deals, cancellations and support tickets, and ask which few items explain most of each. Note anything that has climbed or fallen in the ranking.
Then ask the harder question. Of the hours you personally spent this week, how many went to the few, and how many went to the useful many that someone else could have handled?
Most founders find the answer unflattering the first few times. The number tends to improve once it is being counted.
The bottom line
The 80/20 rule is the Pareto principle, named after an economist who never claimed it by an engineer who later wished he had used his own name instead.
For a startup it is worth far more as a measuring habit than as a motivational quote. Rank your inputs, find the real ratio, protect the few, automate or delegate the useful many, and cut only what is genuinely optional.
Then watch the concentration itself, because the same curve that shows where your growth comes from is also telling you where your business is fragile.
Pick one number this week. Revenue by customer is the usual starting point. Rank it, add up the running total, and find out what your actual ratio is.
