A bubble chart is a set of dots plotted between axes representing two variables. A third variable represents the size of the bubble.
A bubble chart is one of the most popular charts among researchers and analysts. It has a lot of flexibility in terms of variable binding. With some expansions, it can represent up to seven variables at once. At the same time, if not carefully designed, reading bubble charts can be challenging. That’s why it’s better not to overload the viewer's attention by plotting too many variables.
A bubble chart is most commonly used to find correlations. Clusters, as well as outliers, are also easy to spot using bubble charts.
It is a chart with one of the best data/space ratios.
A bubble chart is also known for its versatility. It gives a lot of inspiration to infographic designers and data visualization specialists. It can be turned into almost any chart: heatmap, dot plot, icon chart, tilemap or some hybrid chart.
This chart is very similar to a scatter plot but it’s divided into four equal parts in a 2x2 matrix. It is useful if we want to group data marks for some specific type of analysis (SWOT analysis being one of the best and most well-known examples).
Up to seven variables can be bound in the bubble chart: X- and Y-axes, size, color, stroke color, icon, and label. It’s a record among Datylon charts. But be careful. Too many bound variables can easily confuse the reader. It’s better to limit the number of bound variables to three or four.
Read moreIf an axis type is set to categorical, sorting can be applied: in ascending, descending and reverse order. This can be a huge timesaver if the sorting of categories hasn’t been prepared in the spreadsheet beforehand.
Read moreAll the possibilities of coloring are available in the bubble chart. Three types of color scales allow detailed communication of both categorical and numerical data.
Read moreEvery bubble can be styled separately and tuned to the finest detail. This enables endless options to highlight the bubbles.
Read moreSometimes we don’t even need a bubble. A label on its own can be enough for communicating the message. This is done by hiding the data marks and leaving the labels only. The labels can be colored just like the data marks by using the link option for label color.
Read moreIf you use a combination of numerical and categorical axes and the plots on the numerical axis are dense, you might run into overlapping data marks. To avoid that you can use the Jitter property along the categorical axis. It allows you to spread data marks near the category line and make the data marks more visible.
Read more