How Tableau enables analysts to do their best work (and why you should learn it from me)
Looking to upskill, re-skill, or whatever the hell the corporate world is calling learning something new these days? I have a course for you on Maven!
How I met Tableau and why it blew my mind from day one
Once upon a time, I was a very green data analyst at a healthcare startup. We analyzed data in Microsoft Excel and visualized it in Amazon QuickSight, the world’s worst business intelligence software.
About six months into my tenure there, the company hired a new business intelligence manager who came from a larger company where he’d worked with Tableau. He assessed our data sources, business questions, and current reports and dashboards, then convinced the CEO that Tableau would provide more actionable insights, faster. QuickSight’s user experience and customization options were so poor that I was ready to try any new tool.
The first time the BI manager demoed something in Tableau for me, he didn’t do anything especially fancy, but I was blown away by the speed and ease with which he was able to uncover several key findings by way of exploratory data analysis (EDA). (And we hadn’t even talked about dashboarding yet.) In Tableau, EDA tasks that were simply impossible in QuickSight or would take me an hour in Python could be completed in a few minutes.
The O’Reilly textbook R for Data Science provides this aspirational—and excellent— definition for exploratory data analysis:
EDA is not a formal process with a strict set of rules. More than anything, EDA is a state of mind. During the initial phases of EDA you should feel free to investigate every idea that occurs to you. Some of these ideas will pan out, and some will be dead ends. As your exploration continues, you will home in on a few particularly productive areas that you’ll eventually write up and communicate to others.
Because EDA doesn’t produce a lot of deliverables for stakeholder consumption, data analysts in business settings are almost never afforded the the luxury of enough time to really do EDA this way, even though this approach yields the greatest number of insights and highest quality results.
Fortunately, the power of Tableau greatly reduces the time necessary to do thorough EDA, bringing analysts much closer to a reality where we can actually do the process described in R for Data Science.
For the most part, executives and stakeholders expect actionable insights now (or, if they’re generous, in a few days). Data teams have run around like BI chickens with their graphing heads cut off, meeting unreasonable expectations in unrealistic timeframes for so long that it’s extremely difficult to adjust these expectations to align with the reality of the labor involved.
Most stakeholders have no clue the amount of time, number of surprises, or seemingly endless novel and inevitable—but somehow still unexpected—technical and operational derailments that may crop up in the preparation of data for analysis and visualization, let alone what’s involved in creating engaging, effective, and actionable dashboards. (I experienced an instance of this lack of understanding this week, in fact.) I don’t believe it’s due to a lack of care or interest. They just have their own priorities.
The responsibility for communicating the reality of the time and effort it takes to wrangle, analyze, and communicate data accurately is on companies’ leadership teams. Sadly, many companies don’t have a Chief Data Officer, Chief Information Officer, or other c-suite-level champion for data teams among them. (Chief Technology Officers generally don’t own this role.) Occasionally, another executive will have a background in data, so they can take on the role of data evangelist and advocate for data practitioners in the organization (at one company I worked for, it was the VP of Product, a former data scientist). Until enterprises prioritize becoming truly data-driven rather than just saying that they are, analysts will be expected to produce results fast.
That’s why, if you want to get and keep a real data job in the real data world without burning out in six months, you should learn Tableau. Gartner agrees—check out Gartner’s Peer Insights or G2 Grid for Analytics Platforms. Review some job postings. Many of the highest-paying analyst jobs list Tableau among their must-have skills.
How Tableau allows analysts to do more analysis
Tableau simplifies and accelerates some of the most common time sucks in analysis and visualization, like EDA and data transformation within the tool, while providing absolutely unparalleled options for customization. Want to do k-means clustering analysis? Tableau’s got you. You can apply its algorithm in mere seconds. Want to harness the momentum behind the current AI craze to get your organization more invested in data? Show them how Tableau uses natural language processing (NLP)—a branch of artificial intelligence—to allow them to ask their questions in regular English. They can just Ask Data (introduced in late 2018, not since ChatGPT arrived on the scene).
Tableau makes your work as an analyst manageable, provides you with a sought-after skillset, and helps you deliver results that wow your stakeholders and clients.
How my course is different from other courses
Real data, real problems
Earlier this year, I taught Tableau for another online learning platform. They provided the syllabus, exercises, and datasets the students would be working with. I discovered that they were not teaching students some of the most important skills required to ensure their success. The exercises were unrealistic, the data was pre-cleaned and pre-modeled, and the questions lacked the complexity of real business questions.
I’ve held six full-time data analyst or data visualization developer roles and I’ve worked with dozens of freelance clients. I’m still employed as a full-time senior-level analyst at a large pharmaceutical company in addition to owning my own freelance business. No one has ever handed me a clean dataset with 100 records and asked me to tell them which product sold the most units last month. If they had clean data and easy business questions, they wouldn’t need my expertise.
With Maven, I developed the entire syllabus, hand-selected each dataset, and created challenging exercises based on real projects I’ve worked on in real jobs.
Create a portfolio of work that’s ready to share with potential employers
A portfolio of visualizations of real datasets showcases your problem-solving, data wrangling, communication, analytics, design, and data storytelling skills. Those are the very skills employers want most from candidates, but have difficulty finding.
In my course, with the exception of the first exercise (which is about getting hands-on with Tableau for the first time, connecting to data sources, and working with multiple data sources), we use real-world data with all its imperfections, incompleteness, and untidyness, because the expectation of potential employers is that you know how to handle those datasets. We cover strategies for managing data quality and data modeling issues, which are the most time-consuming part of a data analyst’s job. Using real data leads to discovering real insights which you’ll visualize for your portfolio, and by the end of the course you’ll have completed five portfolio projects you can share via your free Tableau Public profile.
I’m personally invested in the success of Tableau users and data visualization practitioners
I love data visualization and I love Tableau. I believe in it so much that I serve as a Tableau Ambassador and co-lead the worldwide Healthcare Tableau User Group and the local Tampa Bay Tableau User Group, helping people everywhere see and understand data, learn about Tableau’s newest features, troubleshoot issues, share ideas, and—best of all—build data communities. To that end, I founded Women in Dataviz, a tool-agnostic social and professional network for data visualization developers, analysts, and enthusiasts. (Join us on Slack!)
Office hours!
I want to devote the weekly, 90-minute live sessions to teaching you the skills you need to succeed and answering questions related to that specific lesson. (Don’t worry if you can’t make it to a session—they’re all recorded and available on the Maven platform for you to view at your convenience.) However, I know that any given cohort of students will bring varying interests and levels of experience into the course. I want to enable all of my students to reach their data career goals, whether you want to work for a fintech startup, a nonprofit organization, a large healthcare corporation, start your own business, or create beautiful visualizations as a creative outlet or for the enjoyment and education of others.
I expect you’ll be working on your own projects and that you might need help. I want you to be able to ask my advice regarding salary negotiation, freelancing, and interviewing, how to choose the right font or color palette, or how to do linear regression. This is one way Office Hours differentiate my course from others and add the value you’d get from a personal tutor or mentor. If you can’t make it to my regularly scheduled office hours, no problem! We will find a time that works for both of us. I’m committed to your success. When you’re successful, I am successful.
If this is the type of learning experience you’re looking for, I hope you’ll join me on a six-week journey to Tableau Desktop proficiency in my Maven course, Tableau for Data Analytics and Visualization. The sky is the limit once you master these fundamental skills.