Skip to main content
Sankey Chart
K
Written by Konrad Mattheis
Updated over a week ago

Introduction

This article looks at Astrato's Sankey visualization.


A Sankey Chart (alluvial diagram) data object consists of two or more dimensions and one measure.
The Saneky chart gives users a view of data flow between the different dimensions.
Sankey chart insights identify the significance of a path that connects all the dimensions in the chart.

Sankey chart common use cases are:

  • Energy Flow Diagrams

  • Material Flow Analysis

  • Cost Allocation Diagrams

  • Supply Chain and Logistics

  • Website Traffic Flow

  • Population Migration

  • Carbon Footprint Analysis


Building a Sankey

Add a Sankey Chart

The Sankey Chart can be found in the visualization tile. It is positioned in the bottom part of the visualization list.

Define a SankeyChart

  • Add at least two dimensions and one measure.

  • Special Dimension Settings in the Sankey Chart

    • Coloring: set the coloring method for an individual dimension.

      • Single - all nodes in the dimension will have the same color

      • Categorical - each node in the dimension will have a different color.

  • Data Limits

    • By default, the chart displays the first 2000 rows of the query.
      When using more than 3 dimensions, it's recommended to increase the limit of records to a number that covers all the rows.

    • Large number of rows in a charts might impact the performance of the chart and dashboard.

  • Style a Sankey Chart

Layout

  • Show null nodes: show or hide all null nodes across all dimensions.
    Compare the two images below and notice the difference in the second dimension when this setting is enabled versus disabled.

  • Node margin: set the vertical margin between nodes in the dimensions.

Node Labels

  • Show header label: show or hide the dimensions labels.

  • Width: set the width of nodes across all dimensions; when nodes are wide enough, labels will be placed on the side of the nodes.

  • Margin: set the horizontal margin between the links and the nodes.

  • Node labels: enable or disable node labels.
    When enabled, select which information to display:

    • Dimension: only shows the dimension label

    • Flow: only shows the measure value.

    • Both: Showing dimension label and measure value.

  • Placement Priority: by default, the node label is placed automatically in the best position to create a clear display of the chart; by disabling one of the options, you can force a particular place for the node label.

  • Backdrop: set the background color for the node label.

    The image below shows a chart where node labels are set to:
    both, and placement priority is set beside, and the backdrop is light green.

Color

  • Color - set the coloring method for the chart. There are two options:

    • Single: All nodes and links will be colored the same way.
      It's recommended to use this method when wanting to highlight a particular part of the chart.

    • Per Dimension: A different color for each dimension.

    • * This setting is overriding the color setting of the dimensions

  • Conditional coloring: define a coloring condition by either a measure or a dimension.

    • This setting overrides any coloring definition.

  • Node color- when override is enabled, all nodes will be displayed with the defined color.

  • Link color: set the color and gradient mode for links

    • Opacity: set the opacity of links.

    • Override: set the link color, overriding other settings.

    • Coloring method:

      • The left shows a gradient between the start and end dimension colors.

      • The middle option uses the start dimension color

      • The right option uses the end-dimension color


Did this answer your question?