Visualization Options in the Visual Report Builder

Selecting the right visualization for a given data set is a critical piece of the analytical process. Every data set has a story to tell, but the effect of that story is emphasized by its visual impact and readability.

The Magento BI Visual Report Builder offers 12 distinct visualization options, each with their own advantages and use-cases. This article discusses the various visualization options in Magento BI, including required report configurations when applicable, as well as an example of a use-case. The following visualizations are available in Magento BI:

  • Scalar
  • Table
  • Line
  • Bar
  • Stacked Bar
  • Column
  • Stacked Column
  • Pie
  • Area
  • Funnel
  • Scatter plot
  • Bubble

Scalar

Scalar reports are displayed as a single, numeric value. Most often this is used to show the "all time" value of a key metric like revenue or orders, or to compare revenue to date vs budget with two separate scalar reports. In the example below, this simply shows the total number of orders for the given reporting interval:

blobid0

To save a report as a scalar, configure your filters and time settings, then click Save or Update at the top-right of the report. Under the Type drop-down, choose the "Number: Metric name" to save the report as the value shown on the left side-bar.

blobid1

Requirements:

  • Time interval: None
  • Group by: None
  • One metric only

Table

As the name suggests, table reports are great for displaying tabular details. When there is a need to display a large number of group by values or metrics in a single report, a table is often the best way to go. As an example, below is a table of “Customer details”, showing orders and revenue grouped by customer email:

blobid2

Similar to scalar reports, you can save a report as a table by clicking Save or Update within the report builder, then selecting the Table option under the Type drop-down.

blobid3

Requirements:

  • Although there are no report configuration requirements, it’s important to note that tables are limited to 3500 rows. If your data set includes more than 3500 rows, you will need to either filter the results to narrow down the scope, or export the results to CSV or Excel to see the full data set.

Line

Line charts are the perfect choice for comparing the performance of similar metric cohorts. For example, analyzing the revenue of two regions over the same time period, or comparing year over year growth in fulfilled orders, as shown below:

blobid0

Each metric and formula added to the report is represented by its own line. When comparing metrics with similar units and scales, don’t forget to clear the checkbox for “Multiple Y-Axes” to display all metrics on the same scale.

To save a report as a line chart, adjust the report Type to Chart, and select the appropriate visualization from within the report builder, as shown below:

blobid1

Requirements:

  • None

Bar

Bar charts display your data as a series of horizontal bars, and are best for showing overall performance of a limited number of metrics or group by values. For example, a bar chart could be used to compare the revenue by store:

blobid2

Every distinct metric, group by, and time interval combination is displayed as its own bar. If you have two metrics with one group by, containing three distinct group by values, your report will show six separate bars.

To save a report as a bar chart, adjust the report Type to Chart and select the Bar option as shown below:

blobid3

Requirements:

  • None

Stacked Bar

Stacked bar charts are similar to their bar chart brethren, with the additional ability of displaying the proportional breakdown of each bar. Most often, stacked bar charts are set up with two or more metrics and a single group by, such that each bar represents a unique group by value that is split among its metric constituents.

As an example, the report below has two identical revenue metrics: one filtered for first time orders and the other filtered for repeat orders. After grouping by store, you can see both the total revenue contribution for each store (represented by the total width of the bar) as well as the first time vs repeat breakdown of revenue for each store:

blobid4

Make sure the "Multiple Y-Axes" box is unchecked when setting up a report like the above.

To save a report as a stacked bar chart, adjust the report Type to Chart and select the stacked bar option from the report builder:

blobid5

Requirements:

  • None

Column

Column charts represent each data point as a vertical column, and are generally better for displaying time trending data than the horizontal bar chart visualization. Since each unique metric and group by combination is represented in its own series of bars, a column report is generally best for reports with three or less metrics, or one metric with a single group by containing 1-3 group by values.

In the example below, we are showing two revenue metrics, one filtered for first time revenue and the other for repeat revenue, trending over time by month:

blobid6

Column reports can be saved by changing the report Type to Chart, and selecting the column visualization option:

blobid7

Requirements:

  • None

Stacked Column

Stacked column reports are nearly identical to column charts, except similar columns are stacked on top of each other such that the total height represents the sum of the values. Stacked columns are again best visualized with a limited number of metrics or group bys.

Using the same report configuration as described in the Column section above, a report with two revenue metrics (filtered for first time and repeat) would look like the below with a stacked column visualization:

blobid8

Again, it is important that the “Multiple Y-Axes” checkbox is cleared when displaying multiple metrics with the stacked column visualization.

To save a report as a stacked column, set the report Type to Chart and select the stacked column option:

blobid9

Requirements:

  • None

Pie

Pie charts are best for displaying either a single metric with one or more group bys, or multiple metrics with no group bys. In either case, the time interval must be set to none in order to display data in a pie chart. In the example below, a single orders metric is group by store name to show the breakdown of orders by store:

blobid10

To save a report as a pie chart, set the report Type to Chart and select the pie option as shown below:

blobid11

Requirements:

  • Time interval: None
  • Either one of the following:
    • Single metric with one or more group bys
    • Multiple metrics with no group bys

Area

Area charts are almost identical to stacked column charts, except the columns are displayed continuously. Similar to stacked columns, area charts are best visualized with a limited number of group bys or metrics.

Taking the same example from the stacked column section, the report below shows first time vs. repeat revenue with the area chart visualization:

blobid12

To save a report as an area chart, adjust the Type to Chart and select the area option:

blobid13

Requirements:

  • None

Funnel

Funnel charts are perfect for visualizing conversion across an expected sequence of events. A few examples include analyzing the potential revenue in your sales funnel from lead to closed deal, or measuring the drop-off of customers between their first and second orders, second and third orders, etc. An example of the latter is displayed below:

blobid4

In a funnel report, the relative value of a given step of the funnel is reflected by the height of the step, and the order in which the steps are displayed is determined by the report configuration. There are two ways to configure a funnel report:

  • Single metric with one group by: - Order of steps determined by “Show Top/Bottom” setting of the group by. By default, funnel steps are displayed in order from the largest to smallest value, but you can also sort them alphabetically by the group by name.

  • Multiple metrics with no group by: - Order of steps determined by the order that the metrics are added to the report.

To save a report as a funnel chart, adjust the report Type to Chart and select the appropriate visualization from within the report builder.

blobid5

Requirements:

  • Time interval: None
  • Either one of the following:
    • Single metric with one group by
    • Multiple metrics with no group by

Scatter plot

A scatter plot is used to examine a metric’s relationship with two different variables so that you can easily identify correlations and outliers. This type of visualization is best used only with numeric dimensions - try it with the Orders metric and the Customer’s lifetime number of coupons and Customer’s lifetime revenue dimensions to see how coupon usage is related to revenue. You can choose between a scatter plot with and without a trendline:

scatter-plot-1.png

scatter-plot-2.png
Without trendline

scatter-plot-3.png

scatter-plot-4.png
With trendline

Requirements:

Option 1:

  • Two metrics
  • One group by
  • Time interval: None

Option 2:

  • Two metrics
  • No group by
  • Set time interval

Bubble chart

A bubble chart can display up to four dimensions of data where the x and y axes specify the location of the bubbles, the z axis is the size of the bubbles, and by including two group bys you can add color to the bubbles. This type of visualization is best used when you want to plot multiple dimensions of data in a single chart.

For example, the following chart shows the number of customers (bubble size) grouped by a specific acquisition source (bubble color) and state (various bubbles in specific color), plotted against total revenue and average lifetime orders.

bubble-1.png

The following chart shows the number of customers (bubble size) grouped by acquisition source (bubble color) and state (various bubbles in specific color), plotted against average lifetime value and total revenue.

bubble-2.png

Requirements for single series bubble chart:

Option 1

  • Three metrics
  • One group by
  • Time interval: None

Option 2

  • Three metrics
  • No group by
  • Set time interval

Requirements for multi series bubble chart:

  • Three metrics
  • Two group by
  • Time interval: None