Widget Types Reference

Written by Dmitriy Bessedin

Updated on Sep 18, 2026

KPI Scorecard

Learn how to create and configure the KPI Scorecard widget in Improvado dashboards.

Table Widget

Learn how to add and configure the Table widget in Improvado dashboards.

Chart Widget

Learn how to add and configure the Chart widget in Improvado dashboards.

Pie Chart, Funnel, Text Box, and Heatmap Widgets

Learn how to add and configure Pie Chart, Funnel, Text Box, and Heatmap widgets in Improvado dashboards with step-by-step use cases.

Custom Component Widget

Learn how to use the Custom Component widget to embed HTML or React code in Improvado dashboards for bespoke visualizations.

Dashboard Setup and Layout

Step-by-step guide to creating and managing Improvado dashboards, including layout, widget placement, and properties configuration.

Filters, Date Range, and Interactions

Learn how to use filters, date ranges, and cross-widget interactions to explore your data in Improvado dashboards.

Create Dashboards with Detailed Prompts

Ready-to-use prompt examples that help the AI Agent generate accurate, actionable dashboards.

Sankey

Use a Sankey widget to show how volume moves through channels, campaigns, journey steps, or funnel stages. Add it from All widgets under Flows & maps.

Flow between categories: choose one metric and two to four dimensions in flow order, for example Channel → Campaign → Landing Page.
Funnel by category: choose two to six metrics in stage order and, optionally, one dimension such as Channel.

Sankey widgets support dashboard filters and custom SQL. Use Show labels, Show values, Node alignment, and Series limit to control the result. Self-links and non-positive flows are omitted; if the data contains a cycle, the weakest links are hidden so the chart can render.

Geo Map

Use a Geo Map widget to compare one metric by country or US state. Add it from All widgets under Flows & maps, then select a metric and a region dimension.

World accepts country names and ISO alpha-2, alpha-3, or numeric codes.
US states accepts state names, USPS abbreviations, or FIPS codes.
Fit to data focuses the map on regions that contain data; disable it to retain the full map.
Use a Linear scale for evenly distributed values or Quantile when a few regions are much larger than the rest.

Values that resolve to the same region are combined. Unrecognized regions are counted in the legend, which helps identify incompatible names or an incorrect map level.