On this page you'll be presented with the chart types available to the Data Visualization Tool inside the Business Intelligence Module in details and their common applications
Bar charts (and its subtypes: Stacked Bar, Percentage Bar, Column Chart) are best for nominal comparison of individual groups. Shows independence and meaning of each category you have.
You can create a bar chart from the Layout page, click on Chart type and select Bar chart. At least one measure and one dimension are needed to create this chart.
You can add a breakdown dimension to the value axis to create a grouped bar chart. Now you can switch over to a stacked bar chart or use the percentage bar chart to compare elements as parts of the whole.
Finally, to create a chart with multiple metrics, add Measure Names to the category axis. You can check/ uncheck the measures in the filter.
Time series charts (and its subtypes: Timeline, Spline, Step Line, Area, Stacked Area and Percentage Stacked Area) are best for seeing how values change over time, identifying patterns and trends. They are also used to compare groups and show relationship between them over time. Also used to show frequency distribution and deviation over time.
You can create a line chart from the Layout page, click on Chart type and select Time series. At least one measure grouped by time dimension is needed to build a line chart.You can add a breakdown measure to lines (series) to split the measure.
Another interesting option is switching to area charts, which is helpful in visualizing multidimensional data, particularly if stacked. You can only achieve this option by having a breakdown dimension added to the value axis.
Combo charts are very similar to Bar Chart, but with more than one measure. They are useful to compare multiple values across the dimensions set on the category axis.
You can create a combo chart from the Layout page, click on Chart type and select Combo. At least two measures and one dimension are needed to build a combo chart.
Pie and Donut charts
Pie and Donut charts are best for showing the contributions of data segments as a percentage of the whole.
You can create a pie chart from the Layout page, click on Chart type and select Pie. At least one measure and one dimension are needed to create this chart.
You can choose between pie, donut or donut with totals on the Layout tab.
Single Value, Gauge and Progress Bar
Single value is a measure aggregated and presented as a number, while gauge is a measure to compare the featured measure to a related measure. These are used to show, in a glance, a KPI.
You can create a single value card from the Layout page, click on Chart type and select Single value. You need only a single measure for this card.
You can create a gauge or progress bar from the Layout page, click on Chart type and select Gauge or Progress bar. You need a measure you want to analyze and a target measure to establish a comparison for this visualizations. You can format them to show absolute or percentage difference values.
Treemap is appropriate to describe a large amount of information, particularly when you have a dimension with many elements, showing the whole picture as well as how groups are related to each other.
You can create a treemap from the Layout page, click on Chart type and select Treemap. You need a measure to be subdivided into parts and a dimension by which to subdivide it.
You can further group dimensions by color, or add a second measure to compare groups with two measures simultaneously. This option can be enabled on the Format, by clicking in Colors.
Scatter plots are commonly used to find correlations between values (positive or negative, strong or weak).
You can create a scatter plot from the Layout page, click on Chart type and select Scatter plot. You need one dimension and two measures in X and Y axes.
It is also possible to provide more information using the Bubble size option or adding a second measure to compare groups simultaneously.
Funnels (and its subtypes: Pyramid, Single color, Multicolor)are useful to find bottlenecks in a series of successive steps, showing conversion rate from one step to another.
You can create a funnel from the Layout page, click on Chart type and selecting any of the options: Pyramid, Single color, Multicolor. You can either place several measures on the Add measure field (which will be sorted to show correlation) or you can select one measure and define stages by dimension elements.
Maps are the best way to analyze large-scale geographical data. Note that if you do not have any geographical dimension in your data source, this chart type won't appear as an option in the Data Visualization Tool.
You can create a map from the Layout page, clicking on Chart type and selecting Filled (for states maps) or Symbol (for a 'google maps' style of map). You will need a geographical dimension and a measure for this visualization.
There are two levels of map: country (World map) and state. The Data Visualization Tool undestands and creates maps out of data that contains states and countries. The following formats are supported:
- Country names (such as England, France);
- Country codes (on the two-letter uppercase ISO-3166-1 alpha-2 format) or their equivalent English name (such as GB or United Kingdom);
- Region codes (on the uppercase ISO-3166-2 format) or their equivalent English name (such as US-NJ or New Jersey);
- Latitude and longitude data, which generate only the Google Map variant.
Simple Table and Cross-Table
Simple tables are used to show the data the way they are in your data source, without aggregations. Cross-tables work well if you need to emphasize and compare individual numbers.
You can create a simple table or a cross-table from the Layout page, click on Chart type and selecting Simple table or Cross-table. For the simple table, add measures and dimensions as you see fit to create the table columns. For the cross-tables, you need two dimensions (at least one for columns and one for rows) and a measure.
Simple tables do not display the formats defined in the dataset editor level, such as changing numbers to percents and adding a currency symbol.
You can show several measures on a cross-table by placing Measure Names dimension in the columns or rows and setting a filter on the Filter tab. You can also add rows and columns Totals for your cross-table.
You can change the names of the columns of your tables in the Format tab, inside elements, Column headers.
Histograms are used for frequency distribution and continuous data (unlike bar charts, that represent categories variation).
You can create a histogram from the Layout page, click on Chart type and selecting Histogram. You need a dimension and a measure (which will be automatically distributed among bars) to create a histogram
Bullet graphs are used to compare the featured measure to one or more related measures (for example, a target or the same measure at some point in the past, such as a year ago) and relate the featured measure to defined quantitative ranges that declare its qualitative state (for example, good, satisfactory, and poor).
You can create a bullet graph from the Layout page, click on Chart type and selecting Bullet. You need a dimension and a measure to create a bullet graph. Then you can add measures to demonstrate high/middle/low values and a target value (which can be manually set).
Pictorial and Word Cloud Charts
Pictorial and Word Cloud charts are used to represent data comparison visually. These charts use relative sizes or repetitions of the same symbol to show data relation.
You can create a pictorial or word cloud chart from the Layout page, clicking on Chart type and selecting Pictorial or Word cloud. You need a dimension and a measure to create these kind of charts.
Different chart types have different limitations on the number of elements they can display:
- Column and Bar, Combo, Bullet: 300 columns
- Timeline: no limits
- Treemap: 500 segments
- Scatter plot: 1000 elements
- Pie Chart: 1000 segments
- Table: 5000 columns or rows, but no more than 50000 cells.
You can try these alternate approaches to display your data:
- Use another chart type
- Restrict the selection by using chart filters
- Add a scroll to your chart (available for Column and Bar, Timeline and Combo chart types)