Data visualization tools are rapidly becoming popular among small companies or large corporations which allows a company to create a cohesive understanding of their data without having approach to its raw format and with limited knowledge of programming.

Since last post on create a visualize business sales overview report with Power BI Desktop and Tableau Desktop 9.3. In this entry, let’s have a look on the summary with both BI designer tools in 3 areas: Data Connection, Data Preparation and Visualization part. So both tools that we are used for comparison are updated in March 2016.

Data Connections

Data Connections 4

As showed below, both Power BI Desktop and Tableau Desktop 9.3 support a rich set of data connections.

 

Power BI

Tableau Desktop 9.3

Local Files

Excel (*.xls, *.xlsx)

Microsoft Access file

Cube file

Excel (*.xls, *.xlsx)

Microsoft Access file (*.mdb,*.accdb)

Cube file

Text file (*.txt, *.csv, *.xml)

JSON

Folder

Power BI Files

Text file (*.txt, *.csv, *.tab, *.tsv)

Statistical file (SAS,SPSS,R)

Other (Tableau Data Extract files, Tableau workbooks)

Server

Azure - Blob Storage, Data Lake Store (Beta), DocumentDB (Beta), HDInsight, Marketplace, SQL Data Warehouse, SQL Database, Table Storage, HDInsight Spark (Beta)

Tableau Server

Google Analytics

Google Analytics, Google BigQuery, Google Cloud SQL

Hadoop File (HDFS)

Cloudera Hadoop

IBM DB2

IBM – BigInsights, DB2, PDA (Netezza)

MySQL

MySQL

OData

OData

Oracle

Oracle & Oracle Essbase

PostgreSQL

PostgreSQL

Microsoft – Analysis Services, PowerPivot, SQL Server

Microsoft – Analysis Services, PowerPivot, SQL Server

Salesforce – Objects, Reports

Salesforce

SAP – HANA, Sybase

SAP -  HANA, NetWeaver Business Warehouse, Sybase ASE, Sybase IQ

Spark (Beta)

Spark SQL

Teradata

Teradata & Teradata OLAP Connector

Facebook

Web Data Connector

Other Databases (ODBC)

Other Databases (ODBC)

Via ODBC:

Actian Vectorwise

Actian Matrix

Amazon Aurora

Amazon EMR

Amazon Redshift

Aster Database

DataStax Enterprise

Firebird

Hortonworks Hadoop Hive

HP Vertica

MapR Hadoop Hive

Splunk

Actian Vectorwise

Actian Matrix

Amazon Aurora

Amazon EMR

Amazon Redshift

Aster Database

DataStax Enterprise

Firebird

Hortonworks Hadoop Hive

HP Vertica

MapR Hadoop Hive

Splunk

Active Directory

appFigures (Beta)

Blank Query

Dynamics CRM Online

GitHub (Beta)

MailChimp (Beta)

Merketo (Beta)

Microsoft Exchange

QuickBooks Online (Beta)

R Script (Beta)

SharePoint List

Smartsheet (Beta)

SQL Sentry (Beta)

Stripe (Beta)

SweetIQ (Beta)

Twilio (Beta)

Web

ZenDesk (Beta)

EXASolution

MarkLogic

MonetDB

Pivotal Greenplum

Progress OpenEdge

Snowflake

 

Power BI

Power BI supports more cloud based data sources than Tableau. XML file format and properties of the files in a folder are available as data source which are not available in Tableau. There is limitations also, most of the supported database within Tableau although also support in Power BI but users have to figure out if an individual connection is supported or not via ODBC connections.

Tableau Desktop 9.3

Tableau have list the data connections with interface that is easier for users to pick up data source. Tableau also support statistical files in the format of SAS, SPSS and R files. As for R files, although Tableau able to read R files and connect to R through calculated fields which pass values to R via the Rserve package but Power BI able to use R script as it data source which is more straightforward. Tableau does not support Active Directory, Exchange or SharePoint as data source.

Data Preparation

Both Power BI and Tableau take different approach on data preparation. Power BI Desktop provide modeling and transformation in an Excel like interface with a built in Power Query engine where user will be able to transform data in multiple steps within a query window – Query Editor, before user can see how data are presented in report. On the other hand, Tableau allow us to visualize data first then decide any additional integration within data source in a graphic interface or vice versa. Simple features like rename table or column field, transform data type, delete or hide calculated columns are available in both Query Editor and Tableau Desktop’s data connection window.

Power BI

With Power BI, you can merge two queries that use different data sources into one data set and select only the matching rows, remove duplicates, replace one value with another in the selected Columns, transpose rows, pivot/unpivot columns, and parse each cell value in the selected columns as XML or JSON document.

Power BI can also automatically detect the data relationships among queries, you can edit the relationship (1 to 1, Many to 1, 1 to Many) by selecting the columns that relate to one another. But bear in mind that a relationship will not be able to create between two table bacause one of the columns must have unique values.

Tableau Desktop 9.3

In Tableau Desktop 9.3, Tableau Data Interpreter will automatically identify structure of an Excel file. Union features help user to combine Excel or text-based data source which split across multiple files or table into a single Tableau data source. Besides, pivot and split, create groups and bins is allow with selected column in the data grid.

There is also Metadata grid which prepare for user to see all column field’s information at a glance.

Tableau Desktop 9.3 can combine multiple table (from same data source) with available join operator (inner, left, right, full outer) or data blending for blend data from multiple data sources on a single worksheet

Visualization

BI  Tab comparison 3

 

Power BI Desktop

Tableau Desktop 9.3

Chart

Bar chart

Column chart

Stacked bar chart

Stacked column chart

Clustered bar chart

Clustered column chart

100% Stacked bar chart

100% Stacked column chart

Line chart

Area chart

Stacked area chart

Line and Stacked Column Chart

Line and Clustered Column Chart

Waterfall chart

Scatter chart

Pie chart

Tree Map

Map

Table

Matrix

Filled Map

Donut chart

Card

Slicer

Funnel

Multi-row card

KPI

Horizontal Bar Chart

Vertical Bar Chart

Stacked Bar Chart (horizontal)

Stacked Bar Chart (vertical)

Dual combination

Dual combination

100% Stacked bar chart*

100% Stacked column chart*

Line chart

Area chart

Stacked area chart

Dual combination

Dual combination

Gantt bar (with addition custom make)

Scatter chart

Pie chart

Tree Map

Map

Text Table

Text Table

Filled Map

Donut chart *

Text table *

Filter/Parameter function

Funnel *

Text Table *

KPI *

Gauge

Box plot

Bullet Graph

Circle view chart

Gantt chart

Heat Map

Histogram

Packed bubble chart

Summarize

Constant line

Constant line

Average line

Median with Quartiles

Totals

Model

 

Average with 95% Confidence Interval

Median with 95% Confidence Interval

Trend Line

Forecast

Custom

 

Reference line

Reference band

Distribution band

* - (via custom make chart)

Power BI Desktop

Power BI provides visualizations that are very easy to create and modify with just a few drags and drops, switching between different visual charts with just a click. But there is limitation on the customization option as able to rename a title of report, edit axis, but unable to customize the style, font, label or tooltip etc. Although Power BI didn’t provide user the ability to create any custom chart but with Power BI Visuals Gallery which provide user with different data visualization types which created by the Community for the Community. Besides visuals provider from Power BI Visuals Gallery, user also can use R within Power BI Desktop to visualize data by running R scripts.

Tableau Desktop 9.3

Tableau provides rich visualizations with many customization options available. With drag and drop of any data field into row or column shelve or marks shelves, any custom chart can be create by user with Tableau Desktop. Tableau also gave user fully freedom on format sheet, row, column by changing font, align position of font, shading background of sheet, edit border of table or line of a graph.

For mapping options, map are power by Bings within Power BI Desktop and customization are still not available. Whereas in Tableau, due mapping is available and additional data layers on top of a map (street and highways or region borders) is offers to user to add details on map. User can also use map services (WMS server and Mapbox map) which provide in Tableau or import your own custom geocoding.

There are also feature provide by Tableau where user can select desired data points from a visualization or data field as a set for other purpose which this feature is not available in Power BI. Parameter also another tools which enable Tableau user to self-defined the input values within visualizations and still not available in Power BI.

=============================================================================================

Updates

Althought Power BI now have new features where user can create Parameter but the usage of Parameter in Power BI is different compare with the parameter usage in Tableau Desktop. Power BI user can use parameter to filter the data source which user will want.

=============================================================================================

As for calculations part, if you are expert with Excel then you will have no problem to create new measure or new column within Power BI Desktop where all calculations are almost similar in Excel. As for Tableau, either you have some basic programming knowledge or don’t have any programming skill, Tableau provide Quick Table calculation (predefined table calculations: Running total, percent of total, Year over year growth, etc.) or calculated fields (new field created by using existing fields within data) for advanced analysis.

Conclusion

Comparison between Power BI Desktop and Tableau Desktop 9.3 ends here.

I have to mention again that what Tableau Desktop 9.3 can do, Power BI Desktop are able to catch up.

The user-friendly interface of both software allows non-technical users to quickly and easily create simple visual chart to provide insight based on available data.

But, if visualizations are the core focus of your team, Tableau will be the best answer for it. As user are able to create customized dashboard without limit by the available visual charts within software which are different from Power BI Desktop: every visual chart start with a selected visual frame.

Although Microsoft Power BI is a new-comer in the BI realm which it is originally created as a set of add-ons for Excel but it has emerged as a serious contender. Currently, Power BI Desktop are best suited for user who only need simple visualizations or deep reporting capabilities.

Feel free to download both software to explore more with your data:

- Power BI Desktop (free)

- Tableau Desktop 9.3 (free trial up to 14 days)

Thank you for reading and happy analysis!