Box plot visualization (SmartViz)
This feature is in beta and subject to change. Reach out to support with any issues.
The Box plot visualization summarizes the distribution of a measure across categories, showing the median, the interquartile range, and the extremes for each. Use it when an average hides too much, such as latency percentiles across services.
The box plot expects the statistics to already exist as columns in your data. Use Shaping or SQL Analytics to calculate the low, quartile, median, and high values before mapping them here.
When to use a box plot
Reach for a box plot when the spread of a measure matters more than its average. Median, quartiles, and extremes appear together, so a service whose typical latency is fine but whose worst case is not will stand out immediately.
If you have raw values rather than percentiles, use a Histogram, which buckets them for you.
Try it out
Change the settings below and see how the chart updates. Select a dataset from the dropdown to see different use cases.
Response time by service, four boxes of comparable spread. Payments runs both slower and less consistently, and the shape of the box says so on its own.
Box plot settings
Data mapping
X axis
Y axis
Legend & labels
Colors
Add up to 20 colors to set the palette, in order. The palette repeats when there are more items than colors. Leave the palette empty to use the SquaredUp default colors.