How To Use Pareto

usage

slug: example-usage-how-to-use-pareto

Pareto is the official name for the 80/20 rule. This step helps determine at which thresholds x% of a measure is coming from y% of a dimension.

• 80% of revenue comes from 20% of the customers
• 80% of the costs com from 20% of the product purchases
• 80% of support tickets come from 20% of the regions

See the pareto step for more details.

Usage

The pareto step expects its input dataset to be a cube.

Examples

Basic Example

- pareto:
    dataset: salesCube
    measure: revenue
    metric: sum
    output: paretoRanked

Analyze specific level only

- pareto:
    dataset: salesCube
    measure: revenue
    metric: sum
    level: 2
    output: detailPareto

Custom thresholds for aggressive classification

- pareto:
    dataset: productCube
    measure: units
    metric: sum
    paretoThresholds: [70, 85, 95]
    abcBreakpoints: [70, 90]
    calculateGini: true
    output: productPareto

Customer concentration analysis

- pareto:
    dataset: customerCube
    measure: revenue
    metric: sum
    paretoThresholds: [50, 80, 90, 95]
    output: customerConcentration

Chain with filter for detail-level analysis

- filter:
    dataset: salesCube
    where: "level = 2"
    output: detailCube

- pareto:
    dataset: detailCube
    measure: revenue
    metric: sum
    output: detailPareto