5 Traits That Make Great Data Analysts
What Qualities Make Analysts More Likely To Become Experts in Their Craft
A couple of weeks ago I wrote an article about what I call “great data analysts”: those data analysts who know exactly which questions to ask, which database to use with which logic, and can craft super compelling presentations to share their findings.
These data analysts are not born this way - their mother tongue is not SQL, they are not born with a strong business acumen, or with a good understanding of stats. They became great through experience and hard work. But after having interviewed, hired, and worked with XX+ analysts at this point, it seems to me that experience and hard work are mostly catalysts to already existing traits: the skills above are simply the outcome of those traits.
This article that explores those traits is for:
Analytics Managers recruiting data talent.
Data Analysts wanting to improve their skill set.
Anyone curious about a career in data.
#1: Being curious
Curiosity is one of the top qualities for a data analyst (and for a lot of professionals really). By being more curious in life, they are more likely to increase their knowledge, make more connections between ideas, recognize patterns, and, ultimately, be “luckier”. As Sam Altman puts it:
‘Give yourself a lot of shots to get lucky’ is even better advice than it appears on the surface. Luck isn’t an independent variable but increases super-linearly with more surface area—you meet more people, make more connections between new ideas, learn patterns, etc.
Who is more introduced to new ideas or people than a curious person?
A non-curious data analyst might miss a lot of insights that a curious person would uncover. We often refer to the job of a data analyst as being similar to the one of a detective - would you want to work with a detective who is not curious? Granted though - just like most things in life - you need to find the right balance, the one between going too deep into the rabbit hole versus moving on - which is another skill in itself that we’ll talk about shortly.
#2: Thinking logically
Have you ever felt the pain of sitting through a presentation where people jump from ideas to ideas or from assumptions to conclusions without so much proof?
Another top quality for a data analyst is to understand and use logical approaches to problem-solving. Think at minima first-principle thinking and MECE (Mutually Exclusive, Collectively Exhaustive):
First-principle thinking: First-principle thinking is about breaking down the subject matter into its most basic elements, and understanding what are the most important elements to it. In doing so, an analyst will better understand why we are doing what we are doing, and keep the bigger picture in mind. It is really about getting to the crux of the matter - going from the symptoms to the real underlying causes - which more often than not is pretty complicated.
The MECE approach: MECE is about organizing information in a way that every element is separated into distinct sets without overlap (Mutually Exclusive) and together, these sets cover all possible scenarios (Collectively Exhaustive). This approach is extremely helpful for dissecting complex problems into manageable parts - which in turn facilitates deeper insights and more effective solutions. For example, when analyzing sales data, an analyst using MECE might categorize sales into exclusive groups such as "new customer sales" and "repeat customer sales" to ensure a comprehensive yet clear analysis.
Outside of being structured and logical, great data analysts need to be independent and impartial thinkers. Sometimes data can be “weaponized” to justify a narrative; you might have someone very senior sharing with you their thoughts regarding the data before you start your analysis; etc. It is important to stay impartial and to have the courage to report what the data is showing and not showing. I wrote an article a couple of years ago about what to do when an experiment is not returning any significant result - generally speaking, the main ideas can be extended to any data analysis: if your result is not the one that was expected before starting the project, it is important to understand why, to see what can be actionable, and report what you find - just the way it is.
#3: Being Optimist
Another good quality for analysts to have is to be optimistic and confident in their abilities. There are a lot of unknowns with data analyses, and sometimes it can be a bit of an emotional roller-coaster (you find something exciting, you start deep-diving into it, it doesn’t match your initial findings, you ask people around you, you end up realizing that you shouldn’t be using the logic to pull data that you have been using in the first place, etc). It is important to stay optimistic, to have faith in the process, and not to get demotivated and give up too early before finding the right insights.
Similarly, sometimes analysts have to use techniques that they are not necessarily very familiar with, and they have to go back to books & forums on the internet to understand how to perform their analysis (this Reddit thread is pretty interesting on this topic). The best analysts are motivated by this kind of situation as they get the opportunity to expand their knowledge - but this motivation can only come if they are confident (and optimistic) about their ability to learn.
#4: Being Creative
The best analysts can think outside the box and find creative answers to not-so-easy problems. This is a trait that is very linked to curiosity: by being exposed to different ideas and concepts - they can be inspired and come up with their own innovative ideas.
Creativity shouldn’t happen to the detriment of logic though. Imagine your team is working for an online supermarket and you are trying to identify sales patterns - you wouldn’t jump directly into building a complex machine learning model trying to use weather and geopolitical data to predict sales. While this is very creative, a couple of steps might have been missed, and most likely the result won’t be great (that might seem like an extreme example, but this is not far from things that happen in real life).
Creativity is useful at both a micro- and macro-level. On the micro-level, when you have a complicated project and the data is not directly available, someone creative with good coding skills can find alternative ways to work around those constraints. At a more macro level, the ability to think outside the box and approach challenges from different angles can lead to innovative solutions that solve the immediate problem, streamline processes, and improve efficiency overall. This creative mindset, combined with strong analytical skills, allows for the development of unique strategies that can transform data obstacles into valuable opportunities.
#5: Having Good People Skills
People skills are great skills to hone as a data analyst (and quite frankly, those are way too often overlooked). Not only because most people prefer to work with people who have good people skills, but also because it allows you to tap into the institutional knowledge of their organization.
Institutional knowledge is knowledge that hasn’t been documented - or that has been documented but is hard to come by - but lives inside the head of someone inside the company (typical example: which database to use for a specific use case, what studies have already been done about a specific topic, etc.)
Good people skills can also help analysts have uncomfortable conversations (especially when prioritizing projects and pushing back) and insightful discussions with subject matter experts so that they can grow their domain knowledge (which is a very important skill to work on).
Conclusion
In this article, I wanted to deep dive into why great analysts do things differently, and more precisely what is the root cause that made them as good as they are. Hopefully:
As an analytics manager, you have recognized these traits in your best performers and you are clearer with what you should looking for when interviewing someone.
As a data analyst (or someone curious about a career as a data analyst), you are clearer about which traits can help you become expert in your craft.