What you need to know to be a Data Analyst — Technical Skills

Okay, I will skip this intro that working with Data is the future. Companies need more and more data specialists. We are changing our philosophy of management, and an era of Data-Driven Management just came. And you Can be sure that I won’t mention that Harvard Business Review called that Data Specialists is the sexiest job in XXI century. But, really, you won’t hear it here.

What skills do I need to be a successful Data Analyst?

In the beginning, I would like to say that we can find two different groups of skills that are needed. We can split it into Technical Skills and Soft Skills. And I have to say it right — BOTH OF THEM ARE VERY IMPORTANT IN THIS BUSINESS. You cannot be a good analyst with weak, soft skills even when you have excellent technical skills and another way. You can’t be a good analyst with weak technical skills with strong, soft skills.

It is said that you should focus on both of them, but I would say that depends on the level and position where you are. If you are a professional — both of them should be developed equally. On the other hand, if you are starting your career in Data, you can focus on one of them — another one will come with time.

I will focus in this article on Technical Skills.

Technical Skills

Technical Skills are the tools with which you are turning data into valuable insights. I will introduce a couple of them and then try to say how and where you can learn them.

1. Excel, Google Sheets

“What!? Maciej, come on! I want to be a Big Data Analyst. I want to create Artificial Intelligence and create a Skynet, Artificial Brain that will control the world! And you are telling me that the first technical skills that I should work on are Excel or Google? This one which everyone is using in any job!?”

That’s how this world is created. It is standing on glue, tape, and CSV files.

But seriously, Yes, it is essential. I would say that even it is one of the most critical skills sometimes! There are two arguments on that:

• Mostly Excel/Google Sheers is one of the most popular ways to deliver Insights, Visualisations, or just plain data sometimes to your customer/stakeholder.
• It is a straightforward way to understand the basics of Data Science. Honestly, in Excel, you can do everything — you can learn how to prepare data before analysis, keep it simple, clean, calculate essential KPIs, and visualize them. Really.

Never underestimate the power of Excel/Google Sheets!

So what to learn on this topic?

Pivot tables, importing data, visualization, essential functions (sum, mean, and lookup, for example).

Where?

There are plenty of courses on the internet, uDemy, Coursera, or YouTube, so feel free to choose. I recommend this one.

2. Any BI Tool

What to say more — Data Analyst is somehow an artist. Your data need to talk, need to show something. Your shareholders need to understand what you want to show them. I will take care of this topic in the coming posts.

There are many Visualisation Tools, and you will probably meet a couple of them in your career. The three main ones are Tableau, Microsoft BI, and Qlick BI. I am a Tableau fan, and when I will write something about Visualisation Tools, it will be mostly on Tableau.

With Excel/Google Sheet knowledge, you will be able to create the first working Dashboards!

Where to learn something on that?

To learn Tableau, you can find a lot of courses on uDemy and Youtube. Tableau also has its own tutorials on its website. But my favorite course is this one.

3. SQL

In other words — Structured Query Language. It is a language by which you can get your data out of the database. I think that it is fundamental if you want to dive into this “Big Data” sector. For most of the work here, just Excel is not enough. In 99,9%, all of the data analyst work starts in SQL or has any connection with this topic. Also, the well-written SQL query is 80% of Your success.

I recommend here this course: The complete SQL Bootcamp

Also, a tip from my side — in the beginning, you don’t have to focus on the unique dialect of SQL — fundamentals are in 95% the same for all of the dialects.

4. Statistics

One of my biggest mistakes in my career was underestimating the value of statistics in this work. As a self-taught analyst, the biggest shock was when I became an analyst and had problems predicting, statistical tests, etc. These have an enormous value (Am I writing this under every point?). I will touch on the topic of statistics on this blog many times (I will be honest with you — I set this blog mainly for improving my knowledge by trying to learn it ).

Recommended courses:

• Probability and Statistics for Business and Data Science
• Statistics for Business Analytics and Data Science Course
• Designing, Running, and Analysing Experiments

And I also recommend this book: Statistics for Dummies.

5. Python or R

Aaaaaand here we are! After all of these steps, we can come to the programming languages. Most popular in Data Science are Python and R (Scala, Julia, Java are well developed, but in my opinion, learning them could be overkill).

Which one to learn? Python or R? Answer is — It doesn’t matter. Each one of them is excellent for this work. I recommend trying the basics of both of them and then decide. I am a Python user in this case.

Where can I learn to program in R and Python? Of course, I recommend DataCamp as a friendly platform for that, but there is one great course on uDemy on Python that I have to really recommend! Python for Data Science and Machine Learning Bootcamp.

To Sum Up

There are all other technical skills to learn (Git, Schedulers (Airflow, for example), cloud services like AWS, GCP, etc.). Still, these mentioned above are serious fundamentals that will allow you to start your work in the Data Science world or as Data Analyst. I will go deeper into these topics. I will write for sure a post on Soft Skills that are needed, so don’t worry.