2 Proven Coding Languages You Need To Use For Epic Data Visualizations (2022)
This is a great topic, and I love how passionate people are about the coding languages they use! When it comes to prepping your data for visualization though, there are only two you need to learn. In this post I’ll cover each, how it’s used, and help you determine which one to learn first.
The two proven coding languages you need to use for epic data visualizations are 1) SQL and 2) Python. First, SQL is used for the majority of the data transformation and prep work. Then Python is leveraged when more advanced data analysis is required which SQL cannot handle well.
What’s a Coding Language?
A coding language in general is used to interface with data, perform tasks and build applications. By writing text using recognized key words (commands) in a specific way (syntax), you can achieve a desired result. In the example below, the code is telling the database to return all data in a table called “AssociateData” for Brian Barnes. If I use different commands and/or syntax, then I likely will not get the data I need.
There are many similarities between coding languages and traditional languages like English, German and Chinese. In both cases words have a predefined definition and need to be combined in the right way to be meaningful. Using a language with someone who doesn’t speak it means nothing is communicated. Coding languages can also have multiple variations, similar to a spoken language. Have you ever listened to a conversation between people from the US, England and Australia!?
Why Learn a Coding Language?
But why learn how to code? Advanced data prep functionality is being integrated into visualization tools like Tableau and Power BI. There are also tools like Alteryx which specialize in making data prep accessible to people who don’t code. And there are many BI Analysts who simply rely on another team member to prep the data for them. Even with these alternatives available, here are a few reasons why you should still learn to code:
- Codeless Application Limitations. None of the “codeless” applications will give you the full range of capabilities you have through coding.
- Decentralized Logic. Your calculations created in visualization tools are restricted to those tools. Take the case where you need a complex KPI calculation to be displayed in four dashboards. Coding within the visualization tool forces you to replicate that logic four times. Then, each time the logic changes, you have to remember each dashboard which needs to be updated.
- Data sets need to be multi-purpose. Coding allows you to create a data set containing everything you need for visualization. That data set is then available to other visualizations, applications, ad hoc analytics, and even to other data environments.
- Visualization tools change more than coding languages. Different companies use different visualization tools and often swap them out for new ones over time. However, coding languages will rarely change, which makes your skill more valuable over time and across companies.
The 2 Coding Languages You Need to Learn
A quick internet search will return dozens of coding languages, which can be confusing! However, most of those languages perform other functions, such as building applications. Here are the top two languages that are used specifically for data visualization.
SQL
SQL (Structured Query Language, pronounced “sequel” or “S-Q-L”) is a powerful language used to interface with data in relational databases. It has been around since 1974 and has continued to grow in use as the dominant language for managing data.
The BI Analyst uses this language to retrieve source data, add calculations, filter populations and create new data sets. Teams with multiple analysts also use SQL to create base data sets combining key fields needed by entire team. Then, as your data needs grow, SQL will allow you to build out snapshot tables, change history, lookups and more.
The SQL language also has a large variety of variations, which can be confusing at first. As companies started developing database products, they would typically come up with a custom SQL variant to go along with it. For example, I’ve used Microsoft SQL, Oracle SQL, and Teradata SQL over time based on where my data was residing. Most variants are 70-80% the same and if you learn one you can easily pick up others.
Python
The Python language is a multipurpose language with uses that range from application development to data science. For the BI Analyst, our interest is primarily in its capabilities for advanced analytics, which overlaps somewhat with data science. Examples of advanced analytics would be things like forecasting and predictive analytics.
Another advantage with Python is that due to it’s wide use across multiple disciplines, investing your time to learn it will give you more career options. Conversely, spending too much time on niche languages/tools could hurt you in the long run.
Which Language Do You Learn First?
Choosing which one to invest your time on first largely depends on your career objectives. For the BI Analyst, you will generally start with SQL. The SQL language is considered a foundational language with the broadest applications in data prep. If your interest is more along the lines of advanced analytics, then you might want to dive into Python first.
Overall, there is a lot of hype and confusion regarding coding languages. This is primarily because the world of data and technology is evolving rapidly with heavy competition. Hopefully I have convinced you of the benefits in learning a coding language to prep your data for visualization. And if so, then get started with SQL and/or Python, they will unlock many opportunities for you in the future!
