A Day in the Life of a Data Analyst (2023)

In a way, as a

data

analyst

, I am the most important person in the company. Have you ever wondered what a

data

analyst

does on a day-to-day basis? Hi, I'm Tom, and today I'm going to show you exactly that! So what is data analytics and why is it one of the most interesting jobs out there? Data-related roles such as data analyst and data scientist have in the past been called some of the "hottest job titles of the 21st century." So what is a data analyst? Well, there are different ways to answer that question, but in essence, the data analyst must simultaneously be a storyteller, a mathematician, a coder, and a business consultant.

The data analyst uses code to extract data from several different sources and then uses the code again to analyze those data sources to attempt to extract meaning from those data sources and then present those findings, often visually, to the business in general. Today, more and more jobs are remote and a data analyst can basically do their job from anywhere. Personally, I think I like a hybrid model where I spend most of my time working remotely, but still get to a central location where I can be physically face-to-face with my peers and colleagues. That's what I'm going to show you today.

So as a data analyst, I always like to start the day with a nice cup of coffee, so let's do it now. - What is your morning ritual? Big question. My morning ritual varies between meditation and exercise. I try to alternate days between meditation and exercise. - What is your choice of coffee? Oh, what's my choice of coffee? My favorite coffee is a latte with oat milk, so I already have my coffee. I like to spend a few minutes on prep work. So what does prep work mean? Well, I like to spend a few minutes each morning meditating to put my mind in the right space for the day ahead.

I like to review emails from the previous day to make sure I've resolved any issues that are still outstanding. And I also like to check news articles and emails to find out what's new and interesting in the world of data analytics and data science to motivate me for the next day. Now I'm done with my prep work. I would like to spend half an hour in team meetings. Therefore, these meetings are very important for the data analyst role, because often data analysts like to work in team-oriented structures, which means that a data analyst will work with other members of the company or even with other data analysts.

And these team meetings also provide a time-oriented structure. That's why data analysts often like to work in what are known as sprint cycles. That means we like to spend a week or two working toward a particular goal or set of goals. And this morning team meeting is very important so that data analysts can align with other team members on what objectives are crucial to achieving that goal or set of goals. Typically, in that meeting, the data analysts will discuss a backlog of issues that currently exist, and those issues are related to objectives that need to be met, and the data analysts will work with the other team members to prioritize those tasks and such.

Either delete the ones that are no longer important or add new ones. And with the set of tasks prioritized and fragmented, data on this can ensure that day is as goal-oriented and productive as possible. So now that we're done with our team meeting and we know what goals we need to achieve, we'll spend a couple of hours focusing on data analysis. In this data analysis section, we will prepare the tasks. That means breaking down the tasks we've already identified into subcomponents. We will process the data found in the tasks necessary to achieve those goals and we will test and document our processes as we go.

So let's take a specific example. Say, as a data analyst, you're tasked with trying to explain a discrepancy between two different customer segments, and you need to use data from your business to do so. So the first thing you will need to do is extract data from the sources. What does that mean? Well, you'll need to use code like Sequel or maybe Python to extract data from sources, like SQL databases or maybe Excel spreadsheets, to a code repository like a Python IDE, where you can do analysis on the data you you need. I just threw. So now that we have data from our original sources in our IDE code, we can perform exploratory data analysis on that data.

We may be checking to see if there are any data quality issues. For example, we could start by looking to see if any data is missing. We could also check the quality of the data. That is, if we have a name column in our data source, we can check that we don't have numbers in that column. Finally, now that we have data from our sources in our code IDE and performed exploratory data analysis to ensure the quality of the data is high, we will try to answer the business question we were asked using the data at our fingertips.

We can go ahead and close this issue as if it were resolved. Now, all we have left to do is test and document the process. We have an answer to the business question. We need to make sure you are responsible at all times. That's why we created a series of tests that ensure data quality is maintained every time we extract data from its original sources. And we need to document the process so we don't forget what we have done. At each stage of the process we document an explanation, and this will also help us when we start talking to stakeholders to explain the reasons behind the insights we are taking.

Of course, we need to create data visualization and create reports that allow us to communicate our findings to stakeholders. But we've already spent a couple of hours on this. I think it's probably time for lunch. Well, now that I'm back from lunch, and before I get sleepy this afternoon, I want to focus on the main focus of the afternoon. of the way, and that will be producing the reports. There are many ways a data analyst can communicate actionable insights from data, but producing reports is one of the primary ways data analysts will do this. There are typically two ways you'll want to produce reports: interactive dashboards or static presentations.

If you're producing interactive dashboards, you might be using a typical dashboard tool like Tableau, Power BI, or Looker. These are powerful applications that allow you to summarize findings from several different sources and use a variety of different visualization techniques. These powerful applications allow the business to continue analyzing data as the underlying data sources change and are therefore integral to any business requirement. But you can decide to do a presentation instead. These are naturally static, meaning the underlying data doesn't change, so they tend to be used for one-off presentations. When you try to communicate your findings to the company at the end of a project.

Now that you have decided on the format, you just have to go ahead and produce the report. And that basically involves deciding what graphics and what text you are going to include in your reports. Different data sources require different visualization techniques. So make sure you make the right decision about which charts to create. Just as a book with only pictures is boring, and yet a book with only text and no pictures is less attractive. That's why we want to include a combination of graphics and text in our reports. This helps the end user of the company to fully understand the message it is trying to convey.

Well, let's move on to the important topic of code maintenance. If you're a data analyst, you'll almost certainly need to know how to code. That's why I like to spend the next period of my day working on code maintenance. Well, obviously the first component is writing new blocks of code. I like to code in Python and as a data analyst I also use Sequel a lot, but different companies will require different programming languages. So don't be confused if you are asked to learn a new language to join a new company. If you're looking for a new job, don't forget to check the job requirements.

They will often list the coding language they work in or require you to work in. But my personal advice: don't be discouraged if you don't know this coding language yet. Learning a coded language is like learning a foreign language: the more you know, the easier it will be to learn new ones. And people will be impressed if you can already solve problems in their existing code language. So after I've written some of my own code, I like to move on to the peer review process. This is where I, along with the other data analysts in the company, like to work together to test each other's code.

This is an important part of being a programmer because you are not infallible. So having a second pair of eyes reviewing your work not only helps you catch errors, but it also helps you become a better data analyst in the long run. Being a good data analyst also involves writing a lot of tests, but don't worry, they are not as boring as they seem. Writing tests means writing small checks in your code to help you verify, not only that, that the quality of your data is maintained over time, but also that the quality of your code is maintained over time.

So that bugs are not introduced into your code base and so that data quality issues are not introduced into your data sources. Finally, a good data analyst must use version control in their work. This means using platforms like Git or GitHub so that your business logic, the quality of your data, and the quality of your code can be maintained. And if you make a mistake and write a bug in your code, you can always roll back to a previous version. Hey, we've spent quite a bit of time working alone with other data people. Now let's try to focus a little on the business in general.

And for that we are going to organize some stakeholder meetings. Stakeholder meetings are not a constant part of your day. There are some weeks where you will have meetings every day, and others where you will be able to work alone on your code or data all week. In fact, stakeholder meetings are quite atypical. No two stakeholder meetings tend to be the same, but they share some general characteristics. Obviously, communicating your findings or your current work to a wider part of the business and this can range from meeting with people on the product team, meeting with senior management, meeting with finance or even meeting with other data staff . .

In these meetings it is important to have a clear idea of ​​the message you would like to communicate. It is also important to be honest with the business. No one expects you to be perfect and you will encounter problems along the way. So make sure you are honest and transparent with the company. Finally, the role of a data analyst in the company is to help the company make data-driven decisions. So be sure to use the data you've worked on and the reports you've created to get actionable insights and useful suggestions on which direction the business can move forward.

Okay, I'm done with my meeting. I'm going to go back and work some more. Well, at least for today we're done with the stakeholder meetings. So let's move on to the next phase of our day, which is documentation. A good data analyst should always document their work as they go. This helps you remember what you've learned and what you've forgotten so that in the future, when you look back on your work, not only do you know what's been happening, but the entire company knows what's been happening. . All the knowledge you have gained, all the lessons you have learned, and all your findings can be shared with the company at large and saved for posterity.

Okay, it's 5:00. It's not going to be the most productive hour of my day. So instead of doing hard work, I'd like to close my day with some research. It is very important for data analysts to stay abreast of the data analytics, data science and engineering market, because this market changes very quickly. Many new techniques are introduced all the time, so we must spend a good amount of time focusing on research. So where do I go to find good research? Well, from various sources. The first is the first. A good data analyst must be good at mathematics.

So I'll go over the math again and again just to help it stay fresh in my mind. And I can use different mediums for that. I read articles online. I love watching YouTube videos, or I could try to solve some puzzles on aapp on my phone. Yes, we will link all the resources I use in the description below. Another part of being a good data analyst is checking out new code libraries. For example, I'm currently testing a new product called Copilot that helps me write code cleaner and faster. Another important part of being a data analyst is meeting other data analysts, so I like to set aside some time to attend meetups, webinars, meet other data analysts, and network with the broader community.

This is a small glimpse of what a typical day for a data analyst would be like. Obviously some days there are more meetings, other days there is more code, but I hope this gives you a general idea of ​​what a typical day might be like. So if this sounds like something you're interested in, or even if you'd like to learn more, CareerFoundry has a great free short course on data analytics. The link is in the description below. If you enjoy content like this and would like to see more material related to data analysis, be sure to subscribe to the CareerFoundry YouTube channel and click notifications as well.

And if you want to learn more or if something I talked about seems too complex, well, my colleague made a great video, which is an introduction to data analysis. Check here to learn more about what data analysis is, what are the main tasks of data analysis? And basically understand the topic a little better. Check it out and see you soon.