A Sankey Diagram is a visualisation technique that allows to display flows. Several entities (nodes) are represented by rectangles or text. Their links are represented with arrow or arcs that have a width proportional to the importance of the flow.
Here is an example of Sankey…
Here is an example of Sankey…
The Regions, Segments and Categories are depicted as Nodes and the Revenue generated from these nodes are depicted as flows Basically this chart shows the contribution of each region towards the segment and categories in a single chart.
Best Practices..
Limit the position and number of nodes, to improve readability.
Filter out weak connections by allowing for group / having filters in facts.
Provide for Multi select option for nodes – so Business Users Questions are answered.
Adjust data granularity to ensure better response times when displaying and using the Sankey charts
Use Pleasant colors for nodes and connectors so the presentation is pleasing.
Use popups to describe the values in flows – to explain the meaning better.
Click here to know how our Customers use our flexible mentoring service to maximize tableau benefits. Hire an on-demand developer today.. have your business users build sophisticated dashboards and reports..
Line Charts are highly suitable for displaying tendencies and shifts over time; hence a date Field is a prime requirement for this chart type. There are three types of Line Charts,
Continuous Line Chart
Discrete Line Chart, and
Dual Line Chart.
Let’s take a look at each in the mentioned order.
13. Continuous Line Chart
Minimum Requirements:
1 date
0 or more dimensions
1 or more measures
For instance, if you want to show the sales margin of a particular product over time, such as over a year, or two years, or even a span of six months, the Continuous Line Chart will do the work!
You can present the sales margin of more than one product simultaneously by adding more lines that are differently coloured.
Also, adding Trend Lines, (the dotted line shown in the image), to the visualization helps you tell an expansive story quickly.
14. Discrete Line Chart
Minimum Requirements:
1 date
0 or more dimensions
1 or more measures
The difference between the Discrete Line Chart and Continuous Line Chart depends on the type of data fields that are associated with each. The two types of data fields are Discrete and Continuous fields, mapped accordingly to the former and latter chart types.
Continuous fields contain an infinite number of values, while Discrete fields contain finite values.
The Discrete values fall between a range, such as the number of customers in a particular area. Tableau helps you in identifying the fields as Discrete and Continuous, thus saving you the complexity.
The DL Chart enables you to break down the graph for a deeper analysis, to illustrate, take a look at the before-mentioned CL Chart. The graph flows continuously across the selected range of years, while in the DL chart, the graph is sliced in the middle, specifically after four months of each year. The slits in the graph allow you to instantly take a closer look at the sales margin for every four months.
The DL chart works best for analyzing micro-behaviors of large scale data without difficulty.
15. Dual Line Chart
Minimum Requirements:
1 date
0 or more dimensions
2 measures
The Dual Line Chart, also known as the Dual Axis Chart, allows you to present two Measures with two different axes. The Measures are assigned to two axes on either side of the horizontal axis, thereby making it easier to compare two Measures over the same time period.
Do note that the two Measures that you select for comparison should be relevant to each other with a purposeful relationship, for example, Profit and Cost.
Make sure that your selection implies a message to the story that you plan on explaining.
16. Continuous Area Chart
Minimum Requirements:
1 date
0 or more dimensions
1 or more measures
The Area Chart, in general, has the characteristics of both the Line Chart and the Stacked Bar Chart. It helps you to distinctly display the relative proportions of totals or the percentage relationships between Dimensions.
In the event of more than one Dimension, the Area Chart stacks the volume, in colour, beneath the Line. As a result, the total of the fields and also, their comparative size of Measures will be displayed accordingly.
Being a combination of the Line Charts, the Area Chart must contain a date field to create a view over time. In the Continuous Area Chart, the dates are continual.
17. Discrete Area Chart
Minimum Requirements:
1 date
0 or more dimensions
1 or more measures
The Discrete Area Chart is quite similar to the Continuous Area Chart; the only difference is that the former deals with discrete values.
Like the Discrete Line Chart, the Discrete Area Chart allows you to break down the graph as per your requirement.
You can slice a Continuous Area Chart by your desired time periods to take a magnifying look at the involute differences and details.
Area Charts are particularly beneficial for showing visualizations of comparison, such as between product categories, team performances, customer segments, and suchlike. With Area Charts, you can showcase the bigger picture along with the relative magnitude of individual factors.
18. Dual Combination
Minimum Requirements:
1 date
0 or more dimensions
2 measures
The Dual Combination Chart has the layout of a Dual Line Chart, one common axis, the X-axis for the date, and two separate axes, the Y-axes for two different Measures. However, it uses multiple Mark Types for the Measures in the same sheet, hence the word, combination.
A common variant of the Dual Combination Chart is the Line-and-Bar, provided by Tableau.
In the Line-and-Bar Combination type, the first step is similar to creating a complete Dual Line Chart. Once the Second Measure gets dropped into the View, a new Marks shelf will get generated.
Now, the Mark for each Measure can be changed independently of each other, resultantly allowing you to create a View containing a combination of Mark Types.
19. Scatter Plot
Minimum Requirements:
0 or more dimensions
2-4 measures
The Scatter Plot, also known as the Scatter Diagram, or Scatter Chart, or Scatter Graph, is useful to compare two different Measures and to identify patterns.
The Scatter Plot, similar to the Circle View and Side-by-Side Circle View, uses symbols for visualization, which you can change according to your preference.
The distinctive feature of the Scatter Plot is that both the horizontal and vertical axes signify Measures, rather than Dimension. The data points are placed in the graph as specified by the numerical values of both the Measure.
The shape formed by the data points reveals the heart of the story, the correlation present in the entire data. You can drop in Dimensions to add further details to the graph. Also, you can insert Trend Lines to outline the patterns indicating the correlations found in your data.
Scatter Plots show the underlying patterns and relationships existing in a large pool of data, offering an advantage that is unique from the others.
20. Histogram
Minimum Requirements:
1 measure (bin field)
The Histogram, in general, is defined as a graphical representation of the distribution of numerical data.
It is quite similar to the Bar Chart, but the Histogram arranges the values of a Continuous Measure into ranges.
In Tableau, the Histogram creates an additional bin field of your selected Measure, dividing the Measure into discrete intervals, collectively known as Bins.
The Bins help you get a quick analysis of the range of data distribution, also allowing you to add Dimensions to further breakdown the scrutiny.
Such insights will offer you the intricate intelligence needed to improve your business.
21. Box-and-Whisker Plot
Minimum Requirements:
0 or more dimensions
1 or more measures
The Box-and-Whisker Plot, also known as the Box Plot, is used to display the distribution of numerical values along an axis. (Take a look at the image)
Note: In statistical terms, a quartile is a kind of a quantile that divides a data range into four more or less equal quarters.
The Boxes indicate the middle two quartiles of the data’s distribution,i.e., the central 50 percent of the data. The lines jutting out from the top and bottom ends of the box are called the Whiskers. The Whiskers denote the interval between the minimum value of the first quartile and the maximum value of the fourth quartile.
The composition of the graph begins with the user determining the median of the entire data set. The proceedings are as follows,
The vertical intersection in the middle of the Box is the median of the selected data set,the part where there is a color transition from grey to light grey.
The upper and lower quartiles of the Box denote the medians of the upper half and lower half of the data.
The upper half of the Whiskers indicates the maximum of the data set, while the lower half denotes the minimum of the data set.
The Box-and-Whisker Plot works well for descriptive analysis as it presents the shape of the distribution, its central value, and its variability at one glance.
22. Gantt Chart
Minimum Requirements:
1 date
1 or more dimensions
0-2 measures
The Gantt Chart is a great way to depict information relating to time, offering a clear representation of its advancement and the corresponding tasks.
So, it stands as a compelling chart type to explain any progression in time, quite useful when it comes to picturing the various scheduling methods in a Project.
You can take each data point as a task and can plan its correlation and dependencies with other data points and resources. In this way, you get to image the flow of the Project and easily detect potential delays and disruptions.
23. Bullet Graph
Minimum Requirements:
0 or more dimensions
2 measures
You can use the Bullet Graph to compare your present data with the past data or a fixed margin. It is similar to a standard Bar Graph, the difference is, the distribution will be progressing towards a goal behind the Bar in the Bullet Graph.
In general words, the Bullet Graph is used to compare the actual data with a target or previous data, indicated by a reference line. For example, you can compare the present sales figures against the previous year’s sales or the assigned sales target. You can compare the actual cost against the budgeted cost and suchlike.
You can customize the size and colours to make your story more appealing. The Graph can also be presented either horizontally or vertically as per your convenience.
24. Packed Bubbles
Minimum Requirements:
1 or more dimensions
1 or 2 measures
The Packed Bubbles, also known as the Bubble Chart, exhibits relational value in clumps of circles without the use of axes. The individual bubbles denote the Dimensions, and the Measures define the size and colour of the Bubble, that is, the circle.
The bubbles are packed close to each other to save space for additional Dimensions. For instance, while showing the profit and sales percentage of a product, you can easily add different product categories to maximize the analysis.
The Packed Bubbles is a flexible option to display many values concerning one another, for study in a single View. A very colourful View, one can say!
Tableau Desktop, a product of the Tableau Suite, is a powerful data visualization tool that generates actionable insights from extensive data.
As highly competent as it is, Tableau Desktop is relatively easy to use!
With just a few drag-and-drops, you get to have a visual representation of your selected range of data, without intervening time or process.
“Show Me” to Start a View:
“Show Me” comes off as a highly beneficial feature, especially to those who are new to Tableau Desktop.
Select the dimensions and measures that you want to analyze in the Data pane.
Click on “Show Me” in the toolbar to select the type of chart for a visual representation of your data.
Based on the selected dimensions and measures, Tableau evaluates and offers options of charts that you could use for viewing. The available chart types appear highlighted, while the ones considered to be unsuitable for visualization appear more faded to showcase the opposition.
Tableau determines this preference by the number of measures, dimensions, and bins selected in the Data pane.
Note: As you roll over each chart, i.e., the view type, you will be able to see a description at the bottom, displaying the minimum requirements for that particular type.
Types of Tableau Charts:
In total, there are 24 charts in Tableau’s Show Me feature.
Let’s take a look at each of their attributes, purpose, and minimum requirements, one by one.
1. Text Table (Crosstab)
Minimum Requirements:
1 or more dimensions
1 or more measures
The text table, also known as a crosstab, offers a similar View as that of an Excel data source. The mark type is text, with the data displayed in rows and columns, providing a more traditional format of visualization.
However, you can make use of the caption, and colour formatting options to make the text table more appealing. With the mark type being text, you can add extra particulars and information to serve the expectations for a simple yet detailed report!
2. Heat Map
Minimum Requirements:
1 or more dimensions
1 or 2 measures
A heat map provides a more visual approach to showing data compared to a text table. The heat map describes the comparison between categories of data by using color and size as visual cues.
It allows you to convey the underlying context of the data without much verbal explanation, saving you time and effort!
3. Highlight Table
Minimum Requirements:
1 or more dimensions
Exactly 1 measure
A highlight table is easy to create but is also a very compelling chart type. It is a great way to compare value across rows and columns using colour.
The highlight table allows you to apply conditional formatting to a View of your data, in terms of colour. The comparison is displayed either as a continuous or split-levels of colours, ranging from highest to lowest. For example, the higher the sales values, the darker the blue.
You can change the colour scheme and also reverse the sequential order if required. With the highlight table, you can quickly spot the highs and lows and other critical points of activity in your data.
There are exclusively two chart types to choose from when it comes to geographic data; Symbol Maps and Filled Maps.
4. Symbol Map
Minimum Requirements:
1 geo dimension
0 or more dimensions
0 to 2 measures
Symbol Maps, like the name indicates, are maps that use symbols to represent your data. In addition to the default symbols, you can create custom shapes and even change the Marks to different shapes and colours.
Symbol Maps help highlight broad areas of interest and in identifying the concentration of data points in a particular region, such as sales, customers, and suchlike.
5. Filled Map
Minimum Requirements:
1 geo dimension
0 or more dimensions
0 to 2 measures
Filled Maps are quite similar to Symbol Maps; the only difference is that, instead of symbols, you get to fill the geographical region with colour to create a View. You can adjust the transparency and the width of borders to make the visuals more engaging.
Filled Maps offer a quick way to point out and name regions of your data, skipping all the tedious intricate work.
Note: You can also use maps as a filter while working with other types of charts, graphs, and tables. Club together a map with other applicable data, to use it as a filter for a large-scale analysis of your data!
6. Pie Chart
Minimum Requirements:
1 or more dimensions
1 or 2 measure
Whether it’s because of its perfection in shape, or the optical finesse it delivers, pie charts are everyone’s favorite! From classrooms to business presentations!
They are well-suited to displaying proportionate and percentage correlation between data. However, make sure to present the numeric value of the data along with the pie-chart.
For, a View of just the chart without the values can seem ambiguous, in the case of close range numbers.
When used with accuracy, pie charts can swiftly convey the relative value of the data points in the measure.
Note: While using pie-charts, it is advised to limit the number of wedges to six, as opting for a higher number of pie-pieces would make the interpretation and analysis difficult.
7. Horizontal Bar Chart
Minimum Requirements:
0 or more dimensions
1 or more measures
Another popular kid in the neighborhood is the Horizontal Bar Chart. Highly utilized and preferred by a majority of users, this chart type makes information consumption for the viewers quick and easy!
Horizontal Bar Charts display comparative data sets in a clear-cut way, thereby helping in a straightforward and simple analysis. The data sets can be sorted by value to present the highs and lows of a story.
Also, you can apply additional formatting such as by color scheme and borders, to make the View more explanatory.
8. Stacked Bar Chart
Minimum Requirements:
1 or more dimensions
1 or more measures
The stacked bar chart allows you to add additional levels of data sets inside of a horizontal bar measure.
You get to stack another dimension on the data measures of a horizontal bar chart, thereby dividing the entire Measure into corresponding sub-categories. The sub-categories on each bar are then color-coded to exhibit the variation between them.
However, a stacked bar chart is not an ideal choice for comparing individual data points, such as sales by customer. It stands perfect for comparing wholesale data sets, such as overall sales by product, by demographics, and so forth.
9. Side-by-Side Bar Chart
Minimum Requirements:
1 or more dimensions
1 or more measures
The side-by-side bar chart is quite similar to the stacked bar chart, with just one difference. Instead of stacking up the bar segments vertically, they are placed next to each other along the horizontal axis.
You can use this chart type to show a comparison between two or more dimensions or characteristics side by side. For example, comparing the sales volume of a product by regional areas, or comparing the sales volume of a product line at one particular region and so on.
You can format this chart in different ways to highlight the various facets of your story; color-coding is the most commonly preferred among users.
10. Treemap
Minimum Requirements:
1 or more dimensions
1 or 2 measures
Treemaps are the star players of the Tableau Charts, used for illustrating hierarchical data in a tree-structured layout, and part-to-whole relationships.
Their inclusive visual nature makes Treemapping the ultimate choice to display numerous, even thousands of items, in a single View at once. As you mouse over the Nodes in the View, you will be able to see a tooltip appear that displays the specifics of each Node.
The Treemap is quite similar to a heat map, in terms of color-scheme and scalability. However, in Treemaps, the Nodes are grouped by their proximity in the hierarchy.
11. Circle View
Minimum Requirements:
1 or more dimensions
1 or more measures
The Circle View works best for comparative analysis, allowing you to customize the features of the Chart according to your preference.
As the name suggests, the default shape of the Mark is a circle, which you can change to your choice. The color and size of the Mark are decided based on the Keys you determine for the data sets.
For example, the color can point to the Region while the size indicates the Profit, hence the higher the Profit, the bigger the circle!
The Circle View is a persuasive chart type, as it allows you to compare numerous categories simultaneously.
12. Side-by-Side Circle View
Minimum Requirements:
1 or more dimensions
1 or more measures (at least 3 fields)
The Side-by-Side Circle View allows you to add more Measures for displaying a side-by-side comparison, offering an intenser analysis. The Side-by-Side Circle View is a variation of the Circle View, as an extended version.
You can customize the Mark Type to suit the categories set in for comparison, to dilute data overload. This View is a quick way to add additional details within a broader message, enabling you to express your story clearly and concisely!
