Chart Types

Chart.io let's you flexibly analyze your business data using numerous visualizations. Once you've added data to your dataset window (by dragging columns to the X and Y fields), you can use the pulldown menu above the window to select your visualization.

Chartio supports seven types of visualizations. These include:

1. Line chart
2. Bar chart
3. Scatterplot
4. Datetime chart
5. Percent chart
6. Pie chart
7. Table view

Please note: for the time being we do not recommend using Pie charts, as we don't fully support them.

Let's walk through these visualizations one at a time.

Line Chart: As its name suggests, line charts use straight line segments to connect a series of data points.

Line graphs should be used when you want to connect values along an interval scale. Interval scales break up a range of quantitative values into equal quantities, the most common type of which is a time series.

Bar Charts: Bar graphs use rectangular bars with lengths proportional to the values that they represent.

Unlike line charts, bar charts are used for plotting discontinuous data which has discrete values and is not continuous. For example, a bar chart should be used for representing sales by store or region or new unique visits by campaign source.

Scatterplots: Scatterplots are a lot like line charts, but the datapoints are not connected by straight line segments

Whenever time data is involved, we always recommend picking either a bar or line chart. When used properly, scatterplots are ideal for conveying correlation (or lack thereof) between two sets of quantitative values. So, you might want to see whether there is a positive correlation between employee salary and performance review scores, or between sales and a new pricing scheme. The topic of interpreting scatterplots is a bit advanced for this tutorial, so we recommend using these plots only with the appropriate knowledge to accurately interpret them. Here is a link to a Wikipedia article that explains them in more depth.

Percentage Charts: Percentage charts are a great way to visualize ratios between two datasets

Let's say your company has two price plans--premium and free. Using the percent chart type, you can visualize the ratio of these plans among your user base. Create one layer which shows the number of users signing up for the premium plan per week and another layer showing the number of free signups per week and select the percent chart type. This will sum up the values for any given week and then represent each of the datasets as a percentage of the whole. Better than a pie chart, this form of visualization allows you to visualize how ratios change over time. Here, we're showing the ratio of posts and comments on stackoverflow.com.

Table View: Table view is a great way to capture detail in your data

There are a few rules of thumb for when to use a table:

1. You want to look up individual values (say, a list of your top 10 customers)
2. You need to see precise values (say, a comparison of sales revenue across territories)
3. The quantities you're comparing involve more than one unit of measure.

Here we show the number of comments on stackoverflow.com per week.