What's it like working in analytics?
What skills are the most important? What does the day-to-day work look like? What do companies value most?
From my DMs:
How is it working in data analytics, especially in the product and consumer insights field? What skills do you think are most important? What does the day-to-day work look like? What do companies value most from those roles?
I’ve been working in analytics & data science for almost a decade, and while things have changed … honestly, a lot has stayed the same. Including the questions I get asked. So keep scrolling for my answers to some of the most common, all of which I got recently in one DM :)
What are the important skills?
The tools we use haven’t really changed all that much in the past decade.
Excel - Despite all the fancy tools, Excel is still necessary. I know SQL, Python, R, Tableau, and Power BI … and I still use Excel more than you’d expect. If my dataset is small and I don’t need to do much cleaning or more than a simple calculation, a pivot table and a quick visual in Excel is usually quicker than anything else.
SQL - Most companies store their data in some kind of data warehouse, so if you need to use any of it, which is, uh, pretty much every project or task you’ll get, you need to know SQL to get the data you need.
Tableau or Power BI - Every company wants some kind of automated reporting to track their most important metrics with a regular refresh, without asking someone for updated numbers every day/week/month/quarter. A dashboarding tool is still the best way to serve this up.
Arithmetic & descriptive stats - Many roles don’t require much more than addition, subtraction, multiplication, division, mean, median, mode/count, min, max, and quartiles.
Inferential & predictive stats - Not always required, but knowing hypothesis testing, probability, regression, tree-based models, correlations, etc, will open more doors.
Python or R - Not always required, but knowing one of these will open more doors. Python is more common, especially in tech, but R is a great tool and common in more research-based industries.
Business sense - Companies want someone who can take vague questions and problems and use data to solve them. You have to know which follow-up questions to ask, what the data represents, and what metrics matter.
Good communication - You need to be able to translate complex ideas into clear insights and recommendations.
(I’ve been thinking about putting together a course on Communication Skills for Analytics & Data Science. You all will be the first to know if I do. I have a PR Tips for Data course which I can teach again if there is interest - let me know in the comments!)
What about AI? At this point, everyone needs to at least have a basic understanding of generative AI and how to use it - prompting chat, using code companions, and using agents someone else built. And maybe vibe coding, if for no other reason than to be able to see what it spits out and how good or bad it is.
If you work in analytics or data science, you should also have an understanding of how LLMs work and what they’re actually doing. You don’t need to know the math and the intricacies of neural networks, but you should have a general grasp. And if you work in tech, you should have enough knowledge to recommend when or when not to use AI, and how to build (or set up) an agent.
What’s the day-to-day like?
For individual contributors (anyone non-managerial), most roles are 25% meetings and 75% heads down work. This can vary depending on company or team culture and your level of seniority/experience. You’ll have fewer meetings the more junior you are and more meetings as you take on more projects and responsibilities and own your stakeholder relationships. (Meaning you work directly with your stakeholders without your boss supervising everything.)
Meetings are usually team status updates, project syncs, 1:1s with your boss and skip-level manager(s), and company-level all-hands.
Head-down time is time spent doing your tasks and projects.
For data roles in product, customer insights, and marketing, this can include things like
Building a dashboard to track metrics like conversion, retention, churn, new customers, product adoption, ROI or revenue by many different cuts (industry, marketing channel, sales division, country/region, etc)
Understanding what correlates to or predicts churn or retention or product adoption or upgrades or any number of outcomes
Analyzing A/B tests for marketing/sales campaigns or product changes
Calculating customer lifetime value by many different cuts (industry, marketing channel, sales division, country/region, etc)
Writing up an analytics tagging spec or documentation
Here is a longer post about the day-to-day.
What do companies value the most
Being able to answer as many vague questions and solve as many business problems as you can using data.
Your stakeholders and executives don’t really care what method you used, as long as you can take ambiguous questions, fill in the blanks with your domain knowledge and business acumen, and deliver an accurate enough* answer.
*One difference between business and school/Kaggle is that it’s ok to deliver an answer that is 80-90% accurate and move on, versus spending a lot of extra time getting marginal improvements in accuracy. Moving on to another project or task is a better use of time than trying to get an extra 1-2% improvement in your output.
That’s not to say that your boss doesn’t care what tools and methods you know and use. You still need to know how to use a variety of tools and methods so you can pick the right one for your project or task.
And being able to work on your own, without a lot of hand-holding, is very important to your boss, who often has their own work to do. There will be an onboarding period where they need to do some training and provide guidance, even for experienced folks, but after 3-6 months, you should be able to complete tasks on your own. Being a self-starter who can take initiative and solve your own problems is very important if you want to be successful. We’re also often using new tools, and sometimes you will be the first one on your team to have to figure out how to do something. Which can be scary but also fun!
I wrote a post last year on “What are companies actually looking for when hiring for Data Analytics?” and I don’t think much has changed, other than some of them starting to expect you to have a familiarity with AI.
What do you think? Agree or disagree with the above? Let me know in the comments! Have questions of your own? Ask in the comments! You’ll get answers from me and maybe others, and also that’s where I get ideas for future posts.