How to become a Data Analyst in 2023

Data analysis skills are one of the hottest skills that have been in high demand on the job market for the past few years. A "data analyst" job title is not new to the market, however, due to the growth of data generation and the facilitation of data storage provided by cloud computing, many companies have now the capabilities to store their big data and to derive insights and value from it. Data analysis has been and will stay a fundamental skill to have for most jobs. In the following, I will discuss how to start a career as a data analyst and how I was able to secure a job as a data analyst at a reputable company.

Disclaimer

Prepare yourself for the worse; learn more about that here.

You should read it if

You are looking for an internship or a junior opportunity as a Data Analyst.

Data Analyst Trends

A simple search of the term "Data Analyst" on google trends can show us a graph with a positive trend of the frequency of searches. We can observe that from 2004 until 2014 the slope was stable, however, it started growing around 2015 and 2016. In general, the more the term is going to be googled in the coming days and years, the smaller the values will become for the earlier years. Keep in mind that the huge growth can be caused by multiple factors including the new online courses on data analysis, university degrees in data analytics, data analysis jobs ,and much more...

Data Analyst Salary Trends

A much more interesting and more job-related term to observe is "Data Analyst Salary", which can tell us a bit more about how this job is perceived and how many people are curious to know such information. One should keep in mind that the salaries differ from one country to another and from one company to another. Therefore, if you google the term, make sure to be as precise as possible.

What are the responsibilities of a Data Analyst?

A Data Analyst is responsible of transforming the raw data into shapes that would allow or facilitate the analysis of data. In addition, a data analyst can be responsible of performing data visualizations to explore what the data is about and to derive insights from it. Finally, data analysts can perform basic to intermediate modeling of the data similar to what data scientists do, to be able to predict future values.

The transformation part is usually called data cleaning/manipulation. The data visualization part is usually called dashboarding. Finally, the modelling part can be considered as machine learning or predictive statistics... which all refer to the same concept.

Difference between cleaning and manipulation

Data cleaning or data manipulation can be used interchangeably. However, some people consider data cleaning as the transformation that makes the data easier to read and use. For example, cleaning all the empty rows and columns in a dataset or deleting the duplicates... Data manipulation is when we transform the data (pivot, wide to long (see example below), concatenate...) so we can visualize the data. 

An example of a data cleaning and manipulation

An example would be helpful to understand what a data analyst do. Imagine having a receipt from your supermarket but with the bought products written as columns and not rows. It would take a very long piece of paper to write all the products, their amounts and their prices. That's why writing it in a vertical (rows) way makes more sense. This transformation is called 'Wide to long' formatting, it is very popular and is one of the operations that Data Analysts might do.

Data Analyst day-to-day work: example

After finishing the cleaning and manipulation part, a data analyst usually go for the visualization or the dashboarding part. In many cases, you won't be visualizing the transformed raw data on its own, but for example, a comparison between two related datasets. An example would be:

1. Cleaning/transforming the data of sales (multiple products)
2. Comparing the total sales to sales goals set by the CEO of the company
3. Graphing how many products are scoring below or above the goals (thresholds)

The basic skills to become a data analyst

In a few words, I will summarize the fundamental tools to become a data analyst. In addition, I will give some context.

1. Know your Math/Statistics. 
• General math or statistics are important to become a data analyst just so you can make sense of what you're actually doing. You need to know the difference between maximum value, minimum value, mean, median and mode... Additionally, understanding what is the difference between different graphs and their advantages/disadvantages. Most people would've already done these at school, therefore, reviewing some of the concepts won't be a huge amount of work to do.
• The reason why it is important to know some Maths, is to be able to report any "weird" values (or outliers) in your data. As simple as that.
2. Coding skills.
• Do not lose hope, yet! What I mean by coding is basic coding skills. For "Data Analysis" Python is my favorite programming language. Very easy and intuitive, yet very powerful. Many YouTube videos have more than the necessary to proceed.
3. Common Sense.
• A lot of common sense is actually needed to work with data. Knowing when to report data problems is the simplest and most important thing as a Jr./Intern. For example, having a "Height" feature in your data with a mean of 5 meters IS a problem to be reported.

For people trying to become data analysts from scratch (coming from a background with zero math knowledge), it would be very helpful to focus a bit more on understanding the statistics, just the basics. Matter of fact, I have never used advanced statistics in all my internships/jobs. I am not saying that you won't ever use advanced Stat, but as far as you should be concerned, at the beginning, you won't. A beginner's role does not come with lots of responsibilities, which is something normal. It is highly unlikely that a Jr. or Intern will be asked to conduct a fully-fledged "data analysis project" for the company unless there is a shortage of Seniors or the data problem is very simple. In many cases, the Jr/Interns will report back to their manager. Therefore, the focus on simple tasks is going to be the bigger part of the job (cleaning the data and creating basic visualizations).

The intermediate to advanced skills for a data analyst

Now that I have covered the basics, which I believe that many people in 2021 already have acquired, we can jump to the real skills that a data analyst would need for the job. The following would require a good time investment, I would say up to 2 months of daily practice. 

1. Data manipulation and cleaning tools.
• Python or R or Alteryx... As long as you understand one of them, it wouldn't be hard to jump from one to the other. Understanding what a function does is critical, learning how to use it on different platforms/languages isn't. 
• Learning can be done on YouTube, blogs, free courses websites... My go-to was DataCamp. DataCamp is complete but requires a subscription which can be obtained for free for a few months if you're currently a student. DataCamp offers R and Python courses mainly. 
2. Data visualization tools.
• PowerBI, Tableau, Python, R... However, the difference is that Python and R are less flexible when visualizing a certain dataset. On PowerBI and Tableau, one can add slicers to filter certain values, improve the layout... Those are dashboarding programs, in other words, you can use them to showcase a whole data analysis project. 
3. Which visuals to use.
• Understanding what to use for visualization is highly important. One can either learn it on the job or "theoretically". Many books are published on "Storytelling" which can be very helpful, since they contain real life experiences/projects. However, referring to a simple math book is not a bad idea either!
• A summary of all these books: DO NOT USE 3D VISUALIZATIONS! because they are bad (however, pretty).

How I became a Data Analyst

I did my bachelor's in Civil Engineering, where I learned the basics of Python, in the first two years. We learned how to write functions, loops, data structure (high level)... And since I am a Civil Engineer, I believe that I cover the basics pretty nicely.

My advanced statistics were not-top notch, but with a bit of studying here and there, it was not a hard path. My go-to was always StatQuest on YouTube. Josh Starmer, in my opinion, is one of the best teachers out there. He explains the concepts of statistics in a very funny and clear way. He focuses on a diversity of topics (Stat, ML, 'songs' ...) with a nice touch of coding here and there. With that being said, it is highly unlikely that the person will lose motivation or get bored while studying, which is something great!

I started taking DataCamp courses while in my first year of my master's of Artificial Intelligence. I had a free membership sponsored by my university. The courses in my opinion are very well structured. In my opinion, anyone can learn something new on DataCamp, even seniors, because there are always new courses covering new topics. Many people start missing (by skipping) many exercises to receive a certificate, however, for me it was more than that. It was about learning to initiate a new career in a new country. Here's my progress.

Mainly my focus was to be a Data Scientist, however, securing a real Data Science job is not as easy. It requires an in-depth understanding of many concepts that were out of my knowledge due to my background in Civil Engineering. However, there is a lot of overlap between the job of a Data Scientist and a Data Analyst. Many companies use the terms interchangeably and have no idea what they really want to hire, a DA or a DS? And with all the trendy and cool terms that we hear, having a DS title is somehow preferred. 

Finally, the internship that prepared me for my job: I did a four months internship at a Big4 in data analytics. We were mainly using Alteryx. Alteryx is a no-code tool that can be very helpful for data cleaning and manipulation. In addition, it can help a lot in understanding the concepts of merging/joining in tables. Once my no-code cleaning/manipulation skills were rigid, going back to Python was a piece of cake. It became easier to draw an image of what my final output should look like. Moreover, since Alteryx uses buttons as function blocks, one can draw them on paper when thinking about how to tackle and solve a cleaning/manipulation problem.

Can a Data Analyst become a Data Scientist?

Yes, a data analyst can become a data scientist. However, it might take a few years to achieve the new title. A data scientist is in general more knowledgeable than a data analyst in machine learning or models training. This means, for a DA to become a data scientist, the DA has to learn the machine learning concepts, how to use the algorithms and when to use them. In addition, a DA should learn how to tune the hyper-parameters of those algorithms.

Is a Data scientist better than a Data analyst?

Both positions are good and essential for the company. A data analyst can make sure that the data is well prepared for a data scientist to accelerate their tasks. Data scientists can then focus more on the modelling part.

In general, a data scientist can do everything that a data analyst can do.

Am I really a Data Scientist?

Unfortunately, no! Many companies misuse the titles in their job posting to attract people to their company. Not all companies are advanced enough with their data systems to perform real data science. However, since the term data science is over-hyped, companies can sell more of their product if they say that they hire data scientists. Most DS jobs are DA jobs.

Are advanced statistics needed for data analysis?

Advanced statistics are not needed for data analysis, in general. For junior positions, basic to intermediate statistics are more than enough to excel at the job. However, the more you grow within the field, the more you get exposed to advanced concepts. 
To become a data scientist you will need those advanced statistics' concepts, for sure!

How to gain an edge in data analyst jobs?

1. Learn to clean and manipulate data
2. Learn basic modelling algorithms
3. Learn SQL (bonus points)
4. Learn one or multiple dashboarding tools
5. Create a portfolio to highlight your skills
6. Have a master's degree in data analytics

How long it takes to become a data analyst?

It takes around 6 months to become a data analyst from scratch. Learning the basics should take a few weeks. Learning to code in Python or R and all its tools would take 2 months. Learning SQL would take one month. The rest of the days can be used to build a data analyst portfolio. Finally, a couple of weeks and months to apply for jobs and acquire the title.

Who can become a data analyst?

Anyone can become a data analyst. As long as your field of study or work contains data and you believe that the data can be valuable, then you can become a data analyst. Data analysis is a universal tool and should be taught at schools and universities as an essential skill to excel in the job environment.

Thoughts...

It isn't hard. Start working today. Learn the basics. Learn a few advanced stuff. Find an internship.

People reading might wonder why I never mentioned Linux knowledge, SQL... Reason is many companies are not yet too settled. Too many talk about AI and data science/analysis, too little perform any of them. Many companies still get their data as excel sheets which won't require very advanced knowledge of databases...

Read Next 'Is DataCamp Worth It ?'.