A Typology Of Business Charts And How To Use Them
To convince and persuade in today’s corporate world business people must construct evidence-based arguments. They must demonstrate, not simply assert.
Using the right chart during your presentation is essential to demonstrating your point.
Even though everyone is familiar with a few basic business charts, most people aren’t adept at using charts to their fullest data-visualization potential.
Whiz-bang features like animations or sequentially revealed charts are not the answer. Instead, you need to begin with an understanding of what message or point you want your data to communicate, and which type of chart best illustrates this kind of data-driven point.
(Click infographic to enlarge.)
Share this Image On Your Site
Please include attribution to SlideHeroes.com with this graphic
Why It’s Critical To Use The Right Chart
The primary concern while making a presentation is ensuring that your audience will understand the point you are making (the answer to the question you have planted in the mind of the audience).
If your audience does not understand your data or does not follow your presentation closely enough due to disengagement, you will not get your point across to them.
Your job as presenter is twofold:
- Present your data accurately, in a manner that conveys your point or message
- Present your data in a way that makes it easy to understand
Making effective charts goes hand in hand with being able to present those charts effectively.
But specific chart types are better at presenting certain data relationships than others. It is critical to understand what data relationship you are trying to highlight, and what chart type best illustrates this type of relationship.
Why Not Just Use A Table Or A Spreadsheet?
Tables and spreadsheets are frequently an effective means of displaying data.
The primary benefit of a table is that it makes it easy to look up individual values.
The first thing we need to do when trying to determine whether we should use a table or a graph is ask ourselves how the information will be used.
If the primary use will be to look up particular values, then a table is a good option.
Use Tables when you want to use data to:
- Look up individual values
- Compare individual values (but not entire series of values)
- Present precise values
- Present both summary and detail values
Because of their visual nature, graphs present the overall shape of the data.
Use Graphs when:
- The message or story is contained in the shape of the data
- The display will be used to reveal relationships among whole sets of values
Knowing when to use a table and when to use a chart is a matter of understanding the type of data you’re working with, the story you want that data to tell, and the strengths and weaknesses of tables and graphs in presenting data.
Basic Charts And Graphs
Graphs display relationships in quantitative information by giving shape to those relationships.
Quantitative values can be represented in graphs using the following basic tool-set:
- Shapes with varying 2D areas
- Shapes with varying color intensity
What Quantitative Relationship Are You Trying To Illustrate?
There are 8 types of relationships that we use graphs to display.
When determining what type of graph to select, it is absolutely critical that you first consider what you are trying to say with the data.
When you are in the diagnostic phase of your work, you may not know what the data has to say, so you will try a few different approaches.
But once it comes time to creating your presentation, the data on the page exists to support the message you have in your slide headline. You will have a very specific message you will want the data to convey. You will have a specific relationship that you will want to represent.
Let’s review which graph types are most appropriate for each of these 8 types of data relationships.
A nominal comparison graph displays a series of discrete quantitative values to highlight their relative size.
A nominal comparison is best depicted by a bar chart that compares a small number of variables against the same scale.
You can display data either horizontally or vertically, though the standard is to display your data vertically.
A time series is a series of quantitative values that feature something changing over time. Daily closing stock prices over the past month and weekly sales over the last year are examples of time series data.
In a time series, the X-axis represents time points whereas the Y-axis represents the metric of interest.
Line charts do a very good job of displaying the shape of the data over time and are probably the most common time series chart type.
Connecting the individual data points with a line visually links the data and creates an emphasis on the shape of the data, rather than individual data points.
A ranking graph shows how individual values relate to one another sequentially. Which is the largest? Which is the smallest?
Displaying rankings graphically is typically done best with horizontal or vertical bar charts
The key is to sort your data series in ascending or descending order, depending on the main message you want your data to deliver
Part-to-Whole relationships are relationships where each data point is a fraction of a larger complete data set.
Part-to-Whole graphs describe how a part of something relates to the whole. They display the proportion of the whole that each part contributes. Usually it is best to express these proportions as a percentage.
A deviation graph displays how one set of values differs to another primary set of values. Imagine comparing budget figures to actuals, for example.
Deviation relationships are typically arranged in a series of adjacent bar charts, with the difference between the height of each bar in the same data point being the most attention-grabbing.
Distribution graphs show how one or more sets of quantitative values are distributed across their full quantitative range, from lowest to highest.
Bar charts work best, though other options can function just as well (e.g. box and whisker graphs).
Correlation graphs display the relationship between two paired sets of quantitative values to demonstrate whether they are related or not.
The standard way to depict a correlation is via a scatter plot with a line of best fit, which is calculated.
Geospatial graphs display the geographical location of values, positioning values on a map.
Because of the constraints of placing values onto a map, we must rely on size, color, color intensity and width to encode quantitative value.
Select the Right Chart Type to Communicate Your Message
Asserting something is true in a presentation and using the “because I say so’ argument as support (either explicitly or implicitly), is a surprisingly common occurrence.
And make no mistake, it is a mistake.
We need facts. Data.
And we need to present those facts in a compelling, effective manner.
Sometimes neither a table nor a graph will be appropriate. When the quantitative information you want to display consists of a number or two, text will do just fine.
Resist the temptation to sex things up by slapping things into a chart.
But when you have a more complex data-driven conclusion to communicate, you need to use either a table or a chart.
The critical step is understanding which types of charts and graphs are best suited to communicate the data relationship you want to highlight.
- To select between a graph or table, determine how the information will be used
- The primary benefit of a table is that it makes it easy to look up individual values
- Use tables when you want to look up individual values; Compare individual values; Present precise values; Present both summary and detail values
- Graphs present the overall shape of the data
- Use graphs when the message or story is contained in the shape of the data or when the the display will be used to reveal relationships among whole sets of values
- Select the appropriate chart type based on the data relationship being communicated
- “Visualize This” by Nathan Yau
- “The Visual Display Of Quantitative Information” by Edward R. Tufte
- “Show Me The Numbers: Designing Tables And Graphs To Enlighten” by Stephen Few