If you've ever wanted to break into the world of data analytics, but you're overwhelmed by the wealth of concepts and datasets to tackle, this article is for you! We'll walk through:

• Common misconceptions about how to prepare for analytics interviews
• The main skills you should work on

Let's dive in!

Common mistakes in the preparation process

Before we dive into how you should actually prepare, I want to level set here: there are so many things that folks get wrong in preparing for the interview process. Here are a few common mistakes I’ve seen folks make:

• Focusing too heavily on technical skills rather than core concepts and how to reason around data. If you're like I was, then much of your time so far has been focused on mastering technical skills (i.e., SQL, Python, visualization) rather than thinking critically about why and how we use these tools. Don't get me wrong - a minimum technical bar is needed to pass technical screens. But the next steps - in-person interviews, take home assignments - will generally require you to think creatively about the data and communicate your results, not just blindly apply some complex method. In my experience, the unlock here will be based on your ability to:
• Clearly identify and understand the problem you’re trying to solve.
• Define new metrics/features that unravel that problem.
• Communicate your reasoning and your findings.
• Focusing too heavily on visualization tools and capabilities. While this is a subset of the first point, it's worth reiterating specifically for visualizations. I too often see junior candidates build up a portfolio of complex dashboards to show off their visualization skills. This sort of work is largely useless for the job, and interviewers will see right through this. Instead, get good at showing just a line plot or a table so that the takeaway is obvious.
• Practicing with the wrong types of data. I've also seen folks prepare by playing around with a wide, wide variety of public data sets. While this is certainly valuable, it may not be the best use of your time. If you're trying to land a tech job, for example, it's going to be basically useless for the interview process to learn how to work with and think about spatial data and GIS. On the other hand, if you're interviewing at primarily financial institutions, working with clickstream data is not going to help much. It's generally a good idea to decide on a set of industries that interest you first, then prepare incessantly around data for those industries.

Now that those are out of the way, let's get to how you should actually prepare.

Three main categories of skills: data analysis, communication, and wider technical skills

There are 3 main categories of skills required to be a data analyst:

• Data analysis. The actual analysis of data.
• Reasoning & communication. How to turn a business objective into an analysis, then communicate your findings back in a way that makes your findings actionable and accessible.
• Wider technical skills. The world of other skills that might be helpful to know about (e.g. modern data tools, Python) so you can at least talk intelligently about them.

Let's discuss each of these in more depth.

Data analysis: get used to working with data

There's no quick solution here. You'll need to practice, practice, practice. As I've already mentioned, there are a few pitfalls that people commonly fall into when practicing: working with the wrong kinds of data, and diving too deeply into obscure techniques.

The key here is to get used to working with data in the domain that you want to get a job in and in ways that you're likely to encounter during the interview process and on the job. Start by finding some datasets from proximate domains and try to answer some questions yourself. There are tons of publicly available datasets from companies -- just do a quick Google search on a company you're interested in and get going. You need to avoid wasting cycles learning how to parse data that you're not going to encounter.

Practically speaking, play around with distributions, scatterplots, and correlations between variables, products/ratios/other manipulations of variables, and logarithms of these quantities. And understand how to talk intelligently about what these mean. Mostly, that’ll just amount to reason around the mechanisms of correlation and causation between these quantities, but it’s helpful to learn the basics of statistics and statistical reasoning (here’s a great visual guide) to this end as well.

Reasoning and communication: learn to think clearly and communicate precisely

The hard part of analytics work also often isn't the technical work. It's thinking clearly, then communicating your thought process in a convincing way. And if you can’t effectively convey your reasoning, then you won't be to convince anyone to make decisions based on your interpretation of the data. You need to be able to clearly explain your findings and why they matter.

Here are a few strategies I've personally found helpful in developing my reasoning and communication skills, particularly for the interview process:

• Practice talking through your reasoning and results. Talk to yourself about your work, however silly it might seem. This serves two purposes. First, this will force you to fill in any gaps in your understanding/reasoning. Secondly, it'll get you used to talking. You need to train that muscle, as data exploration itself is often entirely intuitive, internal, and unverbalized — it takes effort to give it the right words.
• Read through blog posts published by the company you're interviewing for. Companies will often have blog posts about internal analytics projects, and there will often be a wealth of concepts and ideas specific to your industry within these posts. Familiarize yourself with these concepts, as you’ll likely encounter them during interviews. By learning key jargon beforehand, you won’t have to learn these concepts on the fly and how to think about them during the interview, allowing you to stay focused on the analytics questions at hand.

Wider technical skills: dabble enough in the world of other skills that might be required

You'll want to learn the basics of other skills that you might get exposed to. While you likely won't be interviewed on these ideas, familiarity will give interviewers confidence that your skills will fit their needs. Moreover, hobby-level familiarity with these ideas will help you give off the impression that you're not only interested in the career, but you're truly interested in your domain.

• The modern data stack: having a rudimentary understanding of how modern data stacks work can help you talk intelligently about tooling that might be present in companies.
• Python/R: while generally not a hard requirement of analytics roles, a rudimentary understanding of a scripting language can help you stand out from the crowd.

Conclusion

Analytics can be a lucrative and rewarding career path, but it can be extremely difficult to break into. Focus on understanding and practicing with real data, and you’ll be better prepared than most entry-level candidates. Good luck!