What Data Analysts Do
and Common Myths
What Is a Data Analyst? � � �
The Problem
Every day, the world generates 2.5 quintillion bytes of data (that’s
2,500,000,000,000,000,000 bytes ). To put that into perspective, even if all 8
billion people on Earth filled an Excel spreadsheet to its max of 1,048,576 rows,
it wouldn’t make a dent in that amount of data. That's A LOT of data, and it’s
only growing. For businesses, this data holds the solutions to their biggest
challenges. But here's the catch: data is useless unless someone analyzes it to
uncover those solutions.
The Solution
According to the Cambridge Dictionary, a Data Analyst is:
"A person whose job is to examine information in order to find something out,
or to help with making decisions."
I agree with this definition, but let’s add one crucial clarification:
Your job as a Data Analyst isn’t just to dig into data and "find something
out." It’s to influence decision-makers to make the best possible choices based
on your findings (and this involves meetings… plenty of meetings).
That’s why, as we’ll see today, your role goes beyond just crunching numbers.
So, What Do Data Analysts Do?
1. Identify the Problem
A business stakeholder (i.e., someone who needs your help) will come to you
with a problem, such as: “Why did sales dip last month?”
Typically, this problem is presented in a meeting where the stakeholder
identifies:
Problem To Solve: Understand the issue at hand.
Background on the Problem: Get context for why this problem
matters.
Scope of the Project: Define what is and isn’t included.
Final Deliverables: Know the expected output (e.g., a report,
dashboard, or analysis).
In this phase, you’ll want to ask clarifying questions to ensure you understand
the business impact of solving this problem.
Typically, I’ve had follow-up meetings to clarify whether I’m on the right track,
so it’s rarely just one meeting.
2. Work with the Data
After this, it’s finally time to roll up your sleeves and start putting all those
online
certifications to use.
Now, a data analyst’s job is like working with a big box of LEGO bricks:
Find the LEGO Box: This is called “Data Wrangling.” Locate the data you
need.
Organize the Pieces: The scattered LEGO pieces represent raw,
unorganized data. Clean the data into a usable form.
Group and Explore: Sort the LEGO pieces into groups, akin to
performing “Exploratory Data Analysis” (EDA) to find patterns.
Build the Structure: Assemble all the pieces into a clear, visually
engaging picture—like constructing a LEGO tower to tell a story.
If you get stuck during this process at any time, you’ll be meeting with your
team or stakeholders to get clarification (Yes… even more meetings)
3. Reporting Findings & Repeat
Now that we have our data insights and solutions, it’s time to present our
recommended action to our stakeholders.
Build the Report: Create a presentation or dashboard to visually
showcase your findings.
Present and Communicate: Focus on SHOWING rather than just
TELLING. Visuals and clear narratives make a stronger impact.
Iterate: After presenting, refine your process and prepare for the next
cycle.
Here are three common myths I hear all the time
Myth: You’ll analyze perfectly clean data for your job
When I first started as a Data Analyst, I thought I’d get datasets like the ones
on Kaggle. Boy, was I wrong! The data I got was almost always poor quality—
either incomplete or unusable. I estimate that ~50% of my time was spent just
cleaning datasets to make them usable.
Myth: You’ll hide away in dark caves crunching data away from human
contact
You’ll actually spend a lot of your time talking to other people. This might be in
meetings with your data analytics team, or it could be with business
stakeholders to better understand their needs or present your insights. The
amount of time I spent communicating with others was at minimum a quarter
of my day.
Myth: You’ll need to be a master of advanced mathematics or statistics
The math that I applied as a Data Analyst never got more complex than basic
algebra or descriptive statistics. If you have a high school diploma (or
equivalent), you’ve already learned the minimum math required for this job.
There has never been a time that I (or any fellow analyst I know) have found a
problem they couldn’t solve based on math limitations.