Pareto Law


	[ Origin ] [ Heuristics ] [ Pareto Law ] [ Theory ] [ Examples ]

Pareto Law
The law of the vital few and the trivial many.

Half the money I spend on advertising is wasted, and the trouble is I don't know which half.

Viscount Leverhulme

In the late 1800s, Vilfredo Pareto, an economist, established that 80% of the land in Italy was owned by 20% of the population. This was the first instantiation of a socio-economic law that soon appeared to have universal scope.

The Pareto law, in its generalized form, states that 80 % of the objectives - or more generally the effects - are achieved with 20 % of the means - or more generally the causes or the agents. Subsequently, it takes 80 % of the means to achieve the remaining 20 % of the objectives. In other words, the cost required to move from 80 % to 100 % of the objective is four times bigger than the one required to move from 0 % to 80 %.

Practically, this entails that, beyond a certain threshold, the marginal cost of improving a situation further (e.g. to increase one's market share in a given market segment) becomes prohibitively high.

Hereafter are a few examples where the Pareto law typically applies:

	80 % of a stock is filled with 20 % of the products
	20 % of the customers account for 80 % of the sales volume
	80 % of the profit is achieved with 20 % of the customers (not necessarily the same 20 % as the ones who achieve 80 % of the sales)
	80 % of the wealth is owned by 20 % of the persons (this proportion is highly variable from country to country. It would be interesting to correlate that proportion with the quality of life per country)
	80 % of the traffic pollution is produced by 20 % of the vehicles
	20% of the advertising yields 80% of your campaign's results
	80% of customer complaints are about the same 20% of your projects, products or services

What these figures all tend to point out is that a high level of efficiency (typically 80 %) is achieved with limited means (typically 20 %) but a “near-perfection” efficiency is - relatively speaking - much more expensive.

Those figures are but rough approximations. They all emphasize the highly non-linear distribution of causes and effects or of means and objectives. More often than not that proportion is even exacerbated.

	In the car rental industry the top 0.5 percent of customers rent some 25 percent of cars.
	In the United Kingdom the top 6 percent of cola drinkers drink 60 percent of all colas sold.
	The top 5 percent to 15 percent of U.S. long-distance callers make 55 percent to 60 percent of all long-distance calls.
	At one Midwest bank (in the US), the top 27 percent of customers accounted for 100 percent of profit and covered the losses incurred by the 31 percent of customers on whom the bank was losing money.
	More generally the retail banking sector tends to have a much higher value-skew than other industries : 10 percent of a bank's retail customers can represent 90 percent of its retail profits.
	A few percents of the code of an IT program is sufficient to process almost all situations. The bulk of the code (more than 95 %) is geared toward handling exceptions that but exceptionally occur (in less than 5 % of the cases).

Analytical CRM (Customer Relationship Management) is about finding out which customers, prospects and suspects are among the vital few illustrated by those examples. This requires outstanding application integration and reporting capabilities that few companies actually achieve.

That skew in the contribution to the business benefits might look dreadful but the good news is for almost all businesses one can be as selective as one wants to. A business can skim the most profitable fringe of its market, reap large profits while reducing costs and then expand to another market segment, to another economic sector or to another territory.

Should you be interested in exploiting that leeway to increase the Pareto efficiency of your organization at delivering customer value, do not hesitate to contact us.

Ó IT Cortex


	[ Origin ] [ Heuristics ] [ Pareto Law ] [ Theory ] [ Examples ]