← The collection

Pareto Concentration.

A small fraction can account for a large share when values have a heavy tail.

Interactive experimentintuitiveField note ·
Preparing the experiment…
THE SHORT VERSION

Pareto Concentration, explained.

Pareto concentration describes uneven distributions in which a relatively small share of participants, products, or events accounts for a large share of the total.

01 / THE MECHANISM

Why it happens

A heavy right tail means unusually large values carry substantial weight. The shape parameter controls how concentrated the model becomes. A useful concentration summary depends on the actual distribution; the familiar 80/20 split is a heuristic rather than a guaranteed ratio.

The top-share rule depends on a distribution; it is not always exactly 80/20.

Read the result

Change tail shape and compare the share held by the top group. Look at both the typical observation and the total, because a few large observations can pull them apart.

02 / FOLLOW IT THROUGH

A worked example

Finding a support bottleneck

  1. A team records time spent on support cases and discovers that a small group of cases consumes much of its effort.

  2. It studies those cases separately instead of treating the average case as representative.

  3. The concentration suggests where investigation could help, but does not establish that exactly 20% of cases use exactly 80% of time.

OPTIONAL DEEPER DETAILGo deeper: inside the model

Inside this model

For a Pareto distribution with exponent α=shape/10>1, the top q share is q^(1−1/α). The plot shows the share owned by the top 1–50%.

03 / BEYOND THE EXPERIMENT

Where this idea is useful

A practical use

Compare whether a few products account for most sales in an uneven portfolio.

CHECK YOUR INTUITION

A common misconception

THE TEMPTING CONCLUSION

“Every real dataset obeys the 80/20 rule.”

THE MORE USEFUL DISTINCTION

Concentration is empirical. Some datasets are fairly even; others are much more concentrated than 80/20.

What this explanation leaves out

  • This is one stylized distribution; a measured top share does not prove the underlying mechanism.
ONE MORE QUESTION

Does concentration imply the largest contributors are always the best targets?

No. Their contribution, causes, and cost of changing them matter. A descriptive ranking is not automatically an intervention strategy.

TAKE THE IDEA WITH YOU

Which small group drives the total, and what explains its outsized contribution?

Associated thinkers

Further reading

Explore the original research or the teaching reference behind this experiment.