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

So in a way, as a data analyst, I'm kind of like
the most important person in the company. Ever wondered
what a data analyst does from day to day? 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 coolest jobs around? The roles, around data,
such as data analyst and data scientist have been called in the past some of the
"sexiest job titles in the 21st Century". So what is a data analyst? Well, there are different ways of
answering that question, but at its core, the data analyst needs to simultaneously
be a storyteller, a mathematician,
a coder and a business consultant. The data analyst uses code to pull data
in from various different sources and then uses code again to analyze
those data sources to try and extract meaning from those data sources and then present those findings often
in a visual way to the wider business. More and more jobs
these days are remote working, and a data analyst
basically can do their job from anywhere.

I personally find I like a hybrid model
where I spend most of my time working remotely, but I'm still able
to come into a central place where I can be face to face
physically with my peers and colleagues. So that's
what I'm going to show you today. So as a data analyst,
I always like to start my day with a good cup of coffee,
so let's go and do that now. – What's your morning ritual? Great question. My morning ritual varies
between meditation and exercise. I try and alternate days
between meditation and exercise. – What's your coffee choice? Oh, what's my coffee choice? My favorite coffee is a latte
with oat milk so now I've got my coffee. I like to spend a few minutes
doing prep work. And so what does it mean by prep work? Well, I like to spend a few minutes
every morning meditating to get my mind in the right space for the day ahead.

I like to check emails
from the previous day to make sure I've closed off 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 myself for the day ahead. Now I'm done with my prep work. I'd like to spend half an hour
in team meetings. So these meetings are super important
for the data analyst's role, because often data analysts
like to work in team oriented structures, which means a data analyst will work with other members of the business
or indeed other data analysts.

And these team meetings
also provide a time oriented structure. So data analysts often like to work in
what's known as sprint cycles. That means we like to spend one
or two weeks working on achieving a particular goal
or set of goals. And this morning team
meeting is super important so that the data analysts can align
with the other team members on what objectives are crucial in order
to meet those goal or set of goals. Typically, in that meeting, the data analysts will look at a backlog
of issues that currently exist, and those issues are related
to the objectives that need to be met and the data analysts will work with the other team members
to prioritize those tasks and to maybe remove ones that are no longer
important or add new ones. And with the body of tasks prioritized
and chunked the data on this can then make sure that that day is as goal
oriented and productive as possible. So now that we have our team meeting
out of the way and we know what goals we need to achieve, we're going to spend a couple of hours
focusing on data analytics.

In this data analytics
section, we'll be preparing the tasks. That means breaking the tasks that we've already identified down
into subcomponents. We're going to be processing the data that's in those tasks
required to achieve those objectives, and we're going to be testing
and documenting our processes as we go. So let's take a specific example. Say as a data analyst, you're charged
with trying to explain a discrepancy between two different customer segments
and you need to use data in your business to do that. So the first thing you're going to need
to do is pull data from the sources.

What does that mean? Well, you're going to need to use code
like Sequel or maybe Python to pull data from sources, such a SQL databases
or maybe Excel spreadsheets into a code repository like a Python IDE, where you can do analysis on the data
that you've just pulled. So now that we have data
from our original sources into our code IDE, we can perform
exploratory data analysis on that data. We might be checking
are there any data quality issues? For example, we might start by
looking to see is there any data missing.

We might also check the data
quality of the data. That is that if we have a name column
in our data source, we might check to ensure
that we have no numbers in that column. Finally, now that we have data
from our sources in our code IDE and we performed exploratory data analysis
to ensure that the data quality is high, we're going to try
and answer the business question that we've been given
using the data at our fingertips. We can go ahead and close this question
off as having been solved. Now, all that's left for us to do is test
and document the process.

We have the business question answered. We need to make sure that it's answerable
every time. So we create a series of tests
which ensure that the data quality is maintained every time we pull data
from its original sources. And we need to document the process
so we don't forget what we've done. At every stage in the process we document an explanation, and this will also help us
when we start to speak to stakeholders to explain the reasons behind
the actionable insights that we're taking. Of course,
we need to create data visualization, and we need to create reports
which allow us to communicate our findings to the stakeholders. But we've been spending
a couple of hours on this already I think it's probably time for lunch Well, now I'm back from lunch,
and before I get to sleepy this afternoon, I want to get the main focus
of the afternoon out of the way, and that's going to be producing
the reports.

There are many ways a data analyst can
communicate actionable insights from data, but producing reports is one of the core
ways that data analysts will do this. There are normally two ways you're going
to want to produce reports, either interactive dashboards
or static presentations. If you're producing interactive
dashboards, you might be using a typical dashboarding
tool like Tableau, Power BI or Looker. These are powerful apps that allow you
to summarize findings from various different sources and use a variety
of different visualization techniques. These powerful apps allow the business
to continue to analyze the data as the underlying data sources change,
and so they're integral part of any business requirement. But you may decide to produce
a presentation instead. These are naturally static, which means
the underlying data doesn't change, and so they tend to be used for one off
presentations. When you're trying to communicate at the end of a project,
your findings to the business.

Now you've decided on the format, you just have to go ahead
and produce the report. And that basically involves deciding which charts and what text
you're going to include in your reports. Different data sources require different
visualization techniques. So make sure to make the right decision
about which charts to build. Just like a book
with only pictures is boring, and yet a book with only text
and no pictures is less appealing.

Leave a Reply