I have seen many discussions around Tableau Vs. R. Here are my thoughts on this topic. Tableau is a fantastic tool for pattern discovery using data visualization. It is usually my tool of choice when I want throw some data and keep playing with the data to see whether any patterns emerge. Learning and using Tableau is a very low time consuming activity, but you could keep playing with the data and nothing might emerge.

Whereas, R has a very steep learning curve; any investment you make in R, however, will be returned to you with significant rewards.

R is easily more than a programming language; it is almost a whole framework.

You have access to countless libraries:

• from data manipulation to data visualization 
• from machine learning to financial analysis 
• from interactive Google charts to mining Shakespeare's work 
• to web apps

Why learn R?

If very few people are using R around you, then it is more the reason to learn R, because soon you’ll be wowing them (and future employers) with your skills. Remember: if something is easy to do, everybody is doing that thing; you will become indistinct by becoming a commodity. Better, become indispensable.

Difficulty and learning

There's no sugarcoating this: R is difficult to learn. When you just starting out, you will be confused with matrix and vector, the way R likes to store data.

Hadley Wickham's many libraries, including dplyr and ggplot, and tidyverse make it easy to get started with R.

Here's how Tableau and R compare; notice the distance between them.

As as example of R's power, consider the following report  I created in R.

It is reporting an inclination score (generated using R), showing sparklines (again R), getting data from SQL Server data warehouse, and then repeating the analysis for every region and every capacity range.

More than 100s of nice looking pages within seconds.

If there were only two reasons to use R, I would say these:

1. reproducibility and
2. repeatability.

You can do everything in R in one script. Then you can come back to it after a few years, and still able to track your steps down.

As I see it, it is really not Tableau vs. R issue. They can be used together.

For all Tableau’s promises, I believe that only few people actually use it for exploration and it becomes another reporting a.k.a BI tool.

I use Tableau for exploring, finding quick patterns and then coding that in R to reproduce the patterns for all various combinations. Tableau is not hard to learn at all. You can quickly build dashboards together without actually thinking about the process or about your objective of actionable insights. 

In addition, since it is easy to create charts, there is a real danger of quickly creating many useless charts.

Here’s another example of R’s magic: a convex hull to cluster customers (so getting data, clustering, convex hull and mapping in one single script) to find regions with opportunity.

RStudio provides a nice interface and makes R very easy to use. RStudio supports Markdown which produces nice looking documents or HTML pages with the benefit of retaining all your code to reproduce all your numbers and graphs with newer data.

One more advantage of Markdown: Markdown is very easy to learn.

If you want to create more controlled, beautiful looking documents, however, you can take the complicated, frustrating, yet rewarding route, and rely on knitr, LaTex, Sweave, and RStudio.

More examples of R

Network Analysis

A really good example of the power of R is in creating interactive, web-pages or applications. Here, using the Game of Throne characters and this data set, visNetwork library easily created this fully interactive network. (Note: I'm unable to embed the map in the wordpress easily, but here's an export)

Animation

Using R, I was able to show the growth of Walmart in the US. Graphs, check. Animation, check.

Shiny Apps

One of the biggest advances in R programming was the development of Shiny environment by RStudio. This made creating web-apps very easy. Here's one example from the RStudio gallery:

Don't take my word for it

Here's what some of practitioners and experts in this area said about each Tableau and R: