Our detailed guide to each widget in workflows. Over time, we'll expand each widget into a separate page with a detailed description, visual examples, and comparison to popular BI tools.
Use the text widget to write explanatory notes on your dashboard.
💡 There are no formatting options for the text widget right now.
Use the table widget to display raw data in a grid of rows and columns.
The dashboard user will be able to sort and paginate through the entire dataset.
Use the column widget to display one or more metrics against one dimension.
The horizontal x-axis of the chart shows the Dimension, and on the vertical y-axis are the Metrics. You need to calculate at least one metric, but you can add as many as you want.
📚 Example: Display the CPC and CPM by acquisition channel.
Use the stacked column widget to display one metric against two dimensions.
In a stacked column chart, each of the vertical bars from a column widget is segmented by a second Stack Dimension.
By default, the height of each segment in the bar is the value of the metric. If you choose to Stack 100 percent, the height of each segment is the proportion of the metric.
📚 Example: Display the CPC by acquisition channel and day of the week.
Use the bar widget to display a column widget in a horizontal direction.
Use the stacked bar widget to display a stacked column widget in a horizontal direction.
Use the line widget to display one or more metrics against an ordered dimension (e.g. date, ranking).
The horizontal x-axis of the chart shows the ordered Dimension, and on the vertical y-axis are the Metrics. You need to calculate at least one metric, but you can add as many as you want.
By default, the x-axis is ordered by the Dimension in ascending order. Usually, you'll want to keep it like this as in a line chart, the order of the data points represents something meaningful, like the order of time.
💡 Line charts are currently the best chart for time series data. A dedicated time series chart is coming soon.
📚 Example: Display the CPC and CPM by week.
Use the stacked line widgets to display one metric grouped by a stacked dimension against an ordered dimension.
Use the pie widget to display one metric against one dimension as a percentage.
Each slice of the pie shows the Dimension, and the size of the slice of the percentage of the metric Metric. You can only calculate one metric.
💡 Pie charts are most useful for a few data points with large differences in metrics. For pie charts with more slices and minor variations in the data, consider the column chart.
Use the area widget to display a line widget with a shaded area under the line.
Use the donut widget to display a pie widget, where the slices become segments of a donut shape.
Use the scatter widget to display the relationship between two metrics for one dimension.
The horizontal x-axis of the chart shows the first Metric, and the vertical y-axis is the second Metric. Each data point on the scatter graph shows the values of the two metrics for each category in the Dimension.
The scatter widget is useful to find if any relationship holds between the two metrics, independently of the value of the dimension.
📚 Example: Display the CPC versus CPM by acquisition channel to validate a correlation.
Use the funnel widget to display multiple metrics over the stages of a process.
Each segment of the funnel shows the stage, and the size of the segment is a Metric. By default, the stages of the funnel are ordered by the size of the metric.
📚 Example: Display total users by the stage of an e-commerce purchasing journey.
Use the pyramid widget to display multiple metrics in a hierarchy.
Each segment of the pyramid shows the stage in the hierarchy, and the size of the segment is a Metric. By default, the stages of the pyramid are ordered by the size of the metric.
💡 A pyramid widget is the reverse of a funnel widget.
Use the radar widget to display multiple metrics, optionally against one dimension.
Each data point of the radar chart shows a Metric. If you optionally add a Dimension, each line represents a separate category.
💡 Radar charts are most useful to compare a large number of metrics across a handful of categories. If you only have two data metrics, consider a scatter chart. If you have a large number of categories, consider a column chart.
Use the bubble widget to display the relationship between three metrics for one dimension.
In a bubble chart, each of the data points from a scatter widget has a size determined by the value of a third metric (Size Dimension).
Use the heatmap widget to display one metric against two dimensions, in a grid of rows and columns.
The rows and columns of the table show the two Dimensions, and the value and coloring represents the Metric.
💡 The heatmap widget is like a pivot table with conditional formatting.
A metric widget displays a single numeric value derived from an aggregation. They can be useful to show key insights like KPIs.
Timeseries charts are optimized for working with time data like stock prices. The dimension should always be a time, date or datetime column. Timeseries charts allow you to zoom into particular time frames and bin the data accordingly when getting a high-level view. Currently, we support:
Line and stacked lines
Column and stacked columns
Combination Charts (coming soon 🚀 )
Combination charts allow you to combine line, column, and area charts into a single widget. You can select the secondary Y-axis checkbox to plot a chart on the right Y-axis.