What contributed to changes from one period to the next

business analytics

slug: recipe-business-analytics-what-contributed-to-changes-from-one-period-to-the-next

Recipe: Contribution analysis of period-over-period changes

category: business analytics

Problem

Executives want to know what drove changes between time periods:

• Which segments contributed most to revenue growth or decline?
• What drove quarter-over-quarter changes by product, region, or category?
• Attribution for decision-making and prioritization

Solution

Follow these steps to calculate how various segments contribute to changes from one period to the next:

• ensure the dataset is aggregated by relevant dimensions and measure using cube step
• apply contribution step specifying the compareBy time dimension, measure, and metric
• review contributions by segment
• optionally filter step or [step-rank] to highlight top drivers
• visualize results for executive reporting using print step or chart step

Step Sequence

cube step -> contribution step -> [step-rank] -> chart step

Input Datasets

• sales_cube — multi-dimensional aggregated dataset from cube step
• Notes: must contain compareBy column (e.g., month, quarter) and aggregated measure columns (e.g., revenue.sum)

Output Dataset

• contribution_analysis — dataset with contribution of each segment to total change between periods
• Notes: columns include transition, baseValue, compareValue, contribution, contributionPct, growthRate

Step-By-Step Explanation

Step	Purpose	Notes
cube step	Aggregate metrics across dimensions	Example: total revenue by product × region × quarter
contribution step	Compute change and attribution	Example: compareBy=quarter, measure=revenue, metric=sum
[step-rank]	Optionally rank segments by contributionPct	Identify top drivers of change
chart step	Visualize contributions	Optional waterfall chart, bar chart, or heatmap

Variations & Extensions

• Apply filter step upstream to focus on specific regions, products, or time periods
• Combine with pareto step to identify which segments matter most
• Feed into index step to analyze efficiency vs. volume
• Integrate with dashboards or executive reports

Concepts Demonstrated

• Period-over-period variance decomposition
• Attribution of change to segments
• Integration with multi-dimensional cubes
• Sequencing analytics for actionable insight

Related Recipes

• Multi-dimensional analysis using cubes
• Pareto analysis of contributions
• Index analysis for efficiency

Notes & Best Practices

• Ensure compareBy column is correctly typed and consistent
• Input data should be pre-aggregated to the desired level
• Document measure and metric used for reproducibility
• Always check that contributionPct sums to 100% within each level

Metadata

title: "Contribution analysis of period-over-period changes"
category: "business analytics"
difficulty: "Intermediate"
tags: [contribution, cube, period-over-period, variance]
inputs: [sales_cube]
outputs: [contribution_analysis]
steps: [step-cube, step-contribution, step-rank, step-chart]
author: "Tom Argiro"
last_updated: "2025-10-25"
doc_type: "recipe"