A Day in the Life of a Data Analyst: Myth-Busting

I bet you’ve watched one of those ‘very useful’ YouTube videos, where you’d be expecting to see what a data analyst do on their daily life (AT WORK), but it ends up being ‘the cool life of a data analyst’.

The video starts with a bunch of useless visual effects of the “data analyst” waking up in their half-a-million-dollars bedroom. The best part is that they always wake up with 500 cameras capturing their first seconds of the day, from 500 different angles. They wake up very excited about their day. They are so excited, that the first thing they do is a full 30 mins workout and then they have one of the most luxurious breakfasts you’ll ever see in your life. Finally, and most importantly, they take their shower and right away they start their first ‘business meeting’ where they discuss with 200 executives what they will be doing for the rest of the day.

Sounds familiar? Yes… The sad story is that none of this is real. This is more fictional than the word fiction itself.

How most people wake up for work

No matter how much you love your work, you’re never going to wake up as excited as those YouTube stars. In many cases, you’ll wake up pretty much tired from overthinking the last night due to an issue that you weren’t able to fix -before you call it a day-.

Some days will be so bad, that you will feel like you’ve been hit in the face with a huge brick.

Some other days will also be so bad, that you might wake up 5 mins before your usual working schedule. No breakfast, no brushing teeth, no shower, no 5 minutes stretching before you start your day. Keep in mind that I am assuming that you’ll be working from home; however, some days you’ll have to play the race game with your train that will leave in 5 minutes and you’re 10 minutes away from it. Or you might have to cycle (especially in Europe) so hard to the office to catch up with a meeting, that you might break Armstrong’s cycling record (when he was cycling -slang for taking steroids-).

This is the reality. This is life.

Other aspects of fakeness

Working as a data analyst (junior to experienced), is not as lucrative as people might think it is. Therefore, your bedroom won’t look nice nor have the greatest view ever. You’ll be living in most cases in a normal studio/room.

If you live in Europe, you’d be making around 2000 euros a month. So unless you’re willing to rent a room for a thousand euros… don’t expect much. At the end of the day, the goal is to live with dignity.

The reason why most of these YouTubers might have a fancy life is:

1. Their parents money (rich kids get richer)
2. They are extremely lucky to be working as data analysts in a FAANG (with exceptionally very high salaries)
3. Already making passive income from their YouTube channels
4. Renting a room to film the video… Fake it till you make it… right?

You wouldn’t be eating outside every day, at the most expensive restaurants nor cooking the most expensive steak cuts. Most of your days, you will be eating "at home" cooked food and left-overs. Some other days you will order some fast food (even if you’re very healthy and a gym goer; calories are calories at the end of the day). Exceptionally, or once a week, you will be having those great meals that cost 20+ dollars or euros… I mean, no one is stopping you from doing that everyday… But it shows how financially illiterate you are and how your future investment plans are shit. Finally, your daily meetings won’t be to discuss the companies’ future nor the next two decades strategy. WAKE UP. Most probably the CEO of the company don’t know you exist in their firm. You will most likely discuss your daily tasks with (at most) your supervisor if not your colleagues, and you will be asking/nagging about a certain task that you’re not able to finish due to some difficulties.

This has been my life for over a year in data analysis and data science, at two big 4s, one start-up and one SME (my current job). However, I am not going to lie, I was expecting the life of those YouTubers.

Do not feel bad or wrong

What I said in the previous part might sound very disappointing and discouraging. However, in reality, this is life. You hustle every day.
Sometimes you feel very excited about your days, sometimes you feel very demotivated. The idea is to persevere because at the end of the day, the small battles that you lose temporarily, will make you win the war. By war, I mean your career advancement.
With every bad project, every failure, every project difficulty, you grow.

Data Analyst responsibilities

The responsibilities of data analysts can stretch and vary a lot based on the company they work at. However, some aspects will remain the same no matter what company you decide to join.

You will always:

1. Have new data sets to clean
2. Have new data sets to visualize
3. Check for data inconsistencies (duplicates, outliers…)
4. Check whether you had any information loss while cleaning your data
5. Discuss with your supervisors what are the best practices to solve an issue
6. Discuss with your supervisors what exactly are they expecting of the end product
7. Check again whether the data is still correct after cleaning it
8. Perform unit testing
9. Automate some tasks (if doable)
10. Improve the efficiency of the scripts/functions that you’ve written
11. Learn new tricks or skills. And even whole new concepts (data engineering for example)
12. Raise awareness regarding data

You might:

1. Work on machine learning projects (depends on your skillset)
2. Suggest creative projects that could lead to positive results
3. Take non-critical decisions that wouldn’t affect the project 

You will almost never (I said almost, so you can have some hope):

1. Take critical decisions (remember, we’re talking about juniors or a person with a few years of experience; the data lead will take those critical decisions)
2. Be in direct contact with your client (unless you work within a consulting firm as a data analyst)
3. Work on an end-to-end project on your own and present it to the CEO

An important fact to mention is that a data analyst is not always responsible of doing the analysis of the data. Having technical skills to prepare datasets is not equivalent to having domain knowledge and performing the actual analysis. If you are a CMPS graduate with a master’s degree in data science, this doesn’t mean that you can analyze chemical datasets or sales datasets. You can absolutely prepare the data (better than anyone else) but domain knowledge will come in with exposure and curiosity. You will have to ask the right questions to whom you’re making the dashboard or preparing the data. Certainly, in a few weeks or months, you will be able to have a say in the analysis.

Many people reading this part might find that most of what I said is very intuitive. However, the hype created by social media can lead people to think that if you have a position in data, then you’re a god at the company and no one can mess with you. When in reality, most of the skills that we perform as data practitioners, have been around for years, under uncool names: A STATISTICIAN.

In summary, a DA will clean, manipulate and visualize datasets. A DA will have minor authorities and decision-taking in the company.

An Educator

Like it or not, you will be working as a data-preacher at the company. Otherwise, you will fail miserably at your job. The most important task is to talk about data. This will help your nontechnical colleagues to understand how they should create/collect/request their datasets to make your work a bit easier. For sure they won’t be bringing you perfect datasets; however, having 10 datasets with headers in English, is better than having 10 datasets with headers in 10 different languages that you do not practice. Those minor pre-preprocessing tasks can improve the quality of the time invested as a Data Analyst. Personally, I would rather work on cleaning a dataset to be able to push it to the database and visualize it later on, than work on translating words from Arabic, Portuguese and Russian to English to make sure that my dataset is consistent.

An important task is also to teach nontechnical people about coding. Most people don’t realize that coding is mostly “if else” conditions and that the slightest changes in the data (compared to last week’s data) can make your script throw errors. In general, nontechnical people consider coding as sorcery and magic. They always think that a computer can automatically detect that “Hello”, “hello” and “HELLO” are all the same. It is the case if the algorithm is robust to capitalization, however, in many cases it isn’t. NLP is helping but we’re not there yet and most companies won’t use NLP to improve their cleaning processes.

A final advice I would give is to read the code, out loud, to your non-technical colleagues. Linking the sentences when possible with English words can make it easier for them to understand and to get interested in the field. That way, they will make sure that your dataframe’s headers will always be the same in next week’s data.

A day in the life of an average data analyst

1. Wake up
• Breakfast right away or not
• Brush your teeth right away or later
• Wash your face right away or later
• Have your coffee
2. Turn on the pc
• Check your daily tasks (if you take notes) and start
• Continue a task you’ve been working on
• Having a short meeting with the team or supervisor
• Turn on some music to enhance your mood
3. Start working
• Everything is easy and previously covered (working at ease)
• New difficulties (getting a bit of anxiety)
• StackOverFlow/Google
• Ask your supervisor for help
• Ask friends or use social media (quora, reddit…)
4. Waste some time here and there
5. Midday meeting to report your current progress (might be a message over slack)
6. Have something to eat (normal food for normal humans)
7. Waste some time here and there
8. More focused working
9. A small walk (if you’re at the office) or a coffee break or laying in bed for 10-30 minutes to contemplate life or think of solutions
10. More focused working
11. Somewhere here, you can wrap-up and call it a day

As you might have noticed, you’re not going to work at all times. Even if you’re at the office, you will be wasting some time. Some days will be very busy and 9 hours will pass-by without you noticing. However, an average day should be a bit flexible.