Gentle Introduction To Tableau

Contents

• Introduction to Tableau
• TABLEAU
• Types of Graph in Tableau
• 1. Line Graph
• 2. Bar graph
• a.) Vertical Bar Graph
• b.) Horizontal Bar Graph
• c.) Stacked bar graph
• 3. Heat Map
• 4. Symbol Map and Filled geographical Map
• a.) Symbol map
• b.) Filler Geographical Map
• c.) Heatmap in Geographical form.
• 5. Pie Chart
• 6. Area Charts
• 7. Scatter Plot
• 8. Box-and-Whisker Plot
• 9. Packed Bubbles
• 10. Waterfall graph
• 11. Treemaps

In this post, I am going to introduce you to Data Visualization using Tableau. Data visualization is the visual representation or communication with the help of graphs and plots, dashboards etc. It is a combination of art and science. It is a process where complex data is turned into more accessible and understandable data which can be studied, analysed and questioned by anyone. Today, Data visualization has become the centre of all the field. Whether its stock market, company growth or any field, Data visualizations is the standard of Business Intelligence. Even having sophisticated algorithms and techniques but not be able to convey the insights of the data and result may become the bottleneck of the whole process.

Why Tableau not R or python programming language or any other software such as Power BI.

Well, this is one of the questions I asked myself before starting up with Tableau. I have the experience of working with programming languages such as R and Python. Why should I take additional Burden of learning Tableau just for just data visualization? Tableau is a Data Visualization tool which doesn’t require any knowledge of programming language. In order to use programming languages such as R for data visualizations, one needs to know about the dplyr, plotly and other libraries, similarly in python language, one needs to learn Matplotlib, Seaborn. Anyone working on the tableau can create interactive charts, dashboards, reports and do the statistical analysis like trends and forecasting. When compared to PowerBI, tableau has a lot more functionality and it is available for both Windows and macOS.

TABLEAU

Tableau offers 4 products for data visualization. These are:

1. Tableau Desktop
2. Tableau Prep
3. Tableau Server
4. Tableau Online

These 4 Tableau products come with 14 days of free trial and various plans for the individuals and organizations. For learning purposes, we will use the Tableau Public. Tableau also offers a free license for students Tableau for Students. Download Tableau Public from here. The welcome screen after the tableau install will be:

The left sidebar shows that the type of files that are compatible with the tableau. You can use the Excel, Text file, JSON, Microsoft Access, PDF, Spatial or any statistical file. Further, you can also connect to the server if you or your company is working on. The compatible servers are.

• Tableau Server Data Sources
• Actian Matrix*
• Actian Vector 2.0 or later*
• Amazon Athena
• Amazon Aurora
• Amazon Elastic MapReduce
• Amazon Redshift
• Anaplan
• Apache Drill
• Aster database
• Box
• Cisco Information Server
• Cloudera Hadoop Hive and Impala; Hive CDH3u1, which includes Hive .71, or later; Impala 1.0 or later (incl. Kerberos support for Impala)
• DataStax Enterprise Edition 2.2 or later*
• Denodo
• Dropbox
• EXASOL 4.2 or later*
• Firebird 2.1.4 or later
• Google Analytics
• Google BigQuery
• Google Cloud SQL
• Google Sheets
• Hortonworks Hadoop Hive 1.1 or later
• HP Vertica 6.x or later
• IBM BigInsights*
• IBM DB2 9.1 or later for Linux, UNIX, or Windows (available on Tableau Desktop/Server on Windows only)
• IBM PDA Netezza 4.6 or later*
• JSON files
• Kognitio
• MapR Distribution for Apache Hadoop 2.x or later*
• Marketo
• MarkLogic
• MemSQL
• Microsoft Access 2007 or later*
• Microsoft Azure Data Lake
• Microsoft Azure Data Warehouse
• Microsoft Azure DB
• Microsoft Excel 2007 or later
• Microsoft OneDrive
• Microsoft PowerPivot 2008 or later*
• Microsoft SharePoint Lists
• Microsoft Spark on HDInsight
• Microsoft SQL Server 2005 or later (incl. support for Kerberos)
• Microsoft SQL Server Analysis Services 2005 or later, the multi-dimensional mode only*(incl. support for Kerberos)
• Microsoft SQL Server PDW V2 or later
• MonetDB
• MongoDB BI
• MySQL 5.0 or later
• OData
• Oracle Database 11.0 or later
• Oracle Eloqua
• Oracle Hyperion Essbase 11.1.1 or later*
• PDF
• Pivotal Greenplum 4.x or later
• PostgreSQL 8.3 or later
• Presto
• Progress OpenEdge 10.2B patch 4 or later*
• QuickBooks Online
• Salesforce.com, including Force.com and Database.com
• SAP HANA 1.0035 or later
• SAP NetWeaver Business Warehouse 7.00 with SP20+ recommended; also requires SAP GUI for Windows 7.20 or later client*
• SAP Sybase ASE 15.5 or later*
• SAP Sybase IQ 15 or later*
• ServiceNow ITSM
• Snowflake
• Spark SQL requires Apache Spark 1.2.1 or later
• Spatial files (Esri Shapefiles, KML, GeoJSON, and MapInfo file types)
• Splunk Enterprise 6 or later*
• Statistical Files; SAS (*.sas7bdat), SPSS (*.sav), and R (*.rdata, *.rda)
• Tableau Data Extract
• Teradata V2 R6.2 or later
• Teradata Aster Data cluster 5.0 or later
• Teradata OLAP Connector 14.10 or later*
• Text files — comma-separated value (.csv) files
• Additional databases and applications that are ODBC 3.0 compliant
• Tons of web data with the Web Data Connector
• * are available for Windows only
  Read more here

Types Of Graph In Tableau

In order to proceed with learning Tableau, we need to understand before using the Tableau is the Range of graphs it offers.

1. Line Graph

A line chart or line graph is used to show the change in values over a period of time. The line chart is frequently employed in the stock market where we see the changes in the prices of a stock over a period of time.

The above graph represents the total sale over a span of 4 years and the thickness of the line represents the profit. More the thickness more is the profit.

2. Bar Graph

A bar graph uses horizontal or vertical bars in order to represent data. When the bar graph is used to compare 2 different quantities, it helps the viewer to see the difference in magnitude of the two quantities. Bar graph can help in comparing multiple categories at once against a measured value. It can be grouped and stacked. Group bar graph is a graph where the similar categories are clubbed together. The stacked bar graph is a graph where the different categories are stacked on top of other categories. The stacked graph represents the sales and profit in three different segment.

A.) Vertical Bar Graph

B.) Horizontal Bar Graph

The above two represents the sales in various countries in a vertical and horizontal format.

C.) Stacked Bar Graph

3. Heat Map

Well, we at a certain point of time have acquaintance with the candle flame. A candle flame can be the simple form of a heat map. The Outer Zone is the hottest temperature is covered in the blue and the middle portion is moderately hot and is covered in the yellow and innermost region is the least hot covered in Black.

The same principle is applied in heat maps. There is a change in colour when the category shift from low profit to medium profit or from medium to high profit. The advantages of using heatmap are that the developer doesn’t have to produce numbers in the final graph. This graph is very intuitive. It can be also used to compare 2 quantities by changing the shapes of the blocks with the colours to depict the change in the magnitude of the quantity. Heat maps are quite popular with the corporates. They are readily used among the CEOs, managers meeting.

The heat map represents the magnitude of sales in three different segments across markets around the world. Darker colour` represents the greater profit and vice versa.

4. Symbol Map And Filled Geographical Map

While working with datasets, there may be the case where we have to compare a condition in two quantities across the countries Such as the sale of cars across the globe. Profit of Amazon in different countries etc. Symbol and filled map in tableau help us to plot a geographical plot of our quantities.

A.) Symbol Map

In the symbol map, there is a symbol on the given location in the dataset. We can change the size and colour of the symbol in order to compare different quantities. We can also use pie charts in order to depict the share of these categories in the countries.

B.) Filler Geographical Map

In Filler map, the required location (Continent, Country, State) is covered in colour which can be changed as per the requirement. Further, we can change the colour to create a heat map in order to visualize the change.

C.) Heatmap In Geographical Form.

The heat map can also be used in the filled geographical map. This is helpful in a situation such as comparing the sale of a product across the globe.

5. Pie Chart

The pie chart represents the categorical value or qualitative values. The pie chart is similar to Pizza. Where each slice represents the share/ proportion of the particular category in the total. We can easily analyze that which product is popular according to the proportion of the product in the pie chart.

The pie chart is generally not used when the dataset contains a large number of categories which lead to small proportion and poor visibility.

6. Area Charts

An area chart is similar to line graph in depicting the trends over the time but the key difference is that the area between the graph and the axis is coloured and contains detail which helps in better visualizations and depicts the volume of the categories. Multiple categories can be represented using the stacked area chart. The difference between the graph can be highlighted with the help of colour and size.

7. Scatter Plot

Scatter plot is used to plot the points between the two variables in comparison. The points can be the colour-coded, variable size in order to compare two different categories. It helps in determining the relationship between the variables and to see how dispersed is the dataset. Using different symbols such as triangle square and circle, we can easily compare different categories in a single graph.

8. Box-And-Whisker Plot

Box plot helps in visualization of groups of data using the quartiles. The depicts the 1st quartile, median(2nd quartile), 3rd quartile and two extreme endpoints( Maximum and minimum). It is also called the “5 number summary of any dataset” It helps in analyzing the symmetry of data, how tightly data is grouped. Box plots can be plotted in horizontal and vertical shaped and with colour and size in order to highlight specific category.

9. Packed Bubbles

Packed bubbles is a graph which represents the data in a collection or cluster of a circle. Where the Size and colour of the chart represent the various categories or the types in the graph.

10. Waterfall Graph

A waterfall chart is a form of data visualization that helps in understanding the cumulative effect of sequentially introduced positive or negative values. The initial and final columns can be represented by a full value (100%) whereas the middle columns are initiated where the previous column ends. It can be used to create a quantitative analysis. Using colour-coded columns can help in the visualization of positive and negative effects.

11. Treemaps

Treemaps are a relatively simple data visualization with colour-coded rectangular blocks that help in analyzing and can provide insight in a visually attractive format.

Treemaps are used to visualize the hierarchical structure of the tree Diagram. Treemaps to display data in nested rectangles with each category nested inside of it. Size of the category depends on the magnitude of the category and subcategory. Colours are used in order to differentiate between separate categories.

Tableau can be used to create as many graphs charts dashboards story with just an analytical mind and pick and drop service of the tableau. But it is important to remember that the charts graphs that you create should be appropriate considering the data set. Every graph and chart conveys insights but it is the work of a developer that chooses his/her charts wisely in order to convey the insights effectively and efficiently. Conveying insights doesn’t mean that too many graphs and charts in just one dashboard. These elements create an obstruction in the path of understanding. The person who is analyzing the data may not feel engaged to continue to look at the dashboard when there is too much information on the screen. Why tableau is a combination of art and science. Well, the art is to convey the information in a systematic and beautiful way, science is the method of choosing relevant graphs and charts. This will be the end of the introduction to the tableau. We will be starting off with the basics of the tableau in the coming posts. I hope that through this post you are convinced to continue learning about Tableau and make interesting dashboards and storyboards. Don’t forget to subscribe our website for latest blogs. If you still face any problem or have any doubt? Comment below and we resolve your query as soon as possible.