A short ways back I had the chance to share a little about my career journey in data analytics at a General Assembly event called, Talk Data to Me. And so I wanted to share my story/slides here as well.
First things first, you can access the full deck by clicking the button below, which will take you over to SlideShare (SlideShare seems to be dying a slow death, so if you have any recommendations for other good sites to host presentation decks I'd love to hear it in the comments).
I always think about my presentations/talks as a story, and so this story was broken down into 3 chapters:
Chapter 1 - My Career Path
Chapter 2 - Data, In Real Life
Chapter 3 - Tips & Learning Resources for Aspiring Data Scientists / Analysts
So without further ado, let’s get into it .
Chapter 1: My Career Path
To kick things off I shared a short summary of where I started in my career to where I am today. Everyone's path is going to be different especially in this field), and my key takeaway here was that you should consider this advice when starting off a career in data:
never get too comfortable in a role
always seek out new challenges and problems to solve
always weigh an opportunity to learn and grow over money
This doesn't necessarily mean you need to continuously switch companies, as you might find an employer that keeps you on your toes and constantly throws new opportunity and challenges at you.
In my own experience, I've made several moves over the course of my career, starting out as a data analyst at a digital agency (MRM), to running an analytics/insights team at a communications agency (WE), to running an analytics practice at a tech consultancy (Sapient), and finally to general management with a data provider (YouGov). Each one of these companies offered a new business environment for me to work within as well as vastly different challenges to solve in relation to how businesses can leverage data. Needless to say, each company and role was a new learning experience. Kind of like I was going back to school every time.
But like I said, no two career paths in data analytics will be the same, and moving from company to company isn't for everyone (it can certainly take a toll). Whatever your path, just make sure you’re always being challenged, never get too comfortable and NEVER make a career decision to stay or move somewhere based on money.
Chapter 2: Data, in Real Life
When I was invited to give this talk the organizer asked me to share some examples of how data is used in real-world business scenarios. And so in this section of the presentation I shared two specific examples. One based on the use of data to drive business / digital transformation, and another based on leveraging data to optimize marketing campaign.
The first example, digital transformation, is a story many of you will know, which is Netflix's pivot from being a DVD delivery service to becoming the first and arguably the most successful video streaming platform in the world.
Netflix has always been a data-driven company, from its use of data strategically to inform business strategy (e.g. identifying the opportunity to pivot into streaming) to using data to drive end-user features (e.g.. movie recommendations).
If you're interested in reading about Netflix's various recommendation algorithms, check out this great Quora post. Or if you're looking for a good read about Netflix's pivot from DVD delivery to streaming, you need to read this wonderful post by N. Venkat Venkatraman (Boston University Professor) over on Medium.
Suffice it to say Netflix is company who have built a culture and successful track record around data-driven thinking. And their DVD delivery to video streaming story is a prime example of how companies can leverage data to not only grow but transform their business.
The next example I shared was based on how marketing and communications practitioners use data in everyday scenarios to help them measure and optimize brand or campaign performance.
The chart shown above is just one example of how you can use quantitative research data to measure brand performance and the effect of a crisis. What you're seeing in this chart is brand impression data for Samsung, Volkswagen and United Airlines. More specifically it shows a consumer Impression score - a measure of consumer sentiment which is ranked on a 200 point scale from -100 (very negative) to +100 (very positive) - for the 3 brands in the U.S. over the course of 3 years (January 2015 to May 2018).
Why this chart is so interesting is because it offers a benchmark for how bad the crisis weighed on public opinion in relation to the other brands. Most brands, when engulfed in a crisis, are so focused on getting their brand back on track that they never actually stop to think about how bad the crisis really is. The chart above shows the effect of a major crisis on 3 different brands in the U.S. (i.e. Samsung exploding Note 7's, Volkswagen's emissions scandal, and United Airlines' flight 3411 scandal) and how these compare to each other. You can see from this chart that the United 3411 incident, where airport police forcibly removed passenger David Dao from a United Airlines flight, was by far the most damaging of the 3 examples shown. I mean, who would of thought that beating up your customers could be so bad for the brand 😜
Aside from offering a benchmark, this kind of data can also be useful in terms of informing exactly what kind of PR strategy and messaging are needed to address the crisis. Don't think a measured PR response is important? Just read about Nestlé’s half-billion dollar (and mostly unnecessary) instant noodle recall debacle in India.
Chapter 3: Learning Resources & Tips for Aspiring Data Scientists/Analysts
In the third and final section of the presentation I offered 5 tips and resources that I think may be of help to aspiring data scientists or analysts.
Tip #1 - Get Inspired
This is by no means limited to a career in data analytics, but an important characteristic of a fulfilling career, I believe, is that you constantly seek out and immerse yourself in things or people that inspire you. Knowing where to go to get your a dose of inspiration can help keep you engaged in your role as well as help you expand your mind and discover new ways to problem solve.
I shared a list of 14 influencers who have served as a great source of inspiration for me over the years, including the late Hans Rosling, Edward Tufte, Nancy Duarte and many more. I won't go through them 1 by 1, but you can check out the full list on SlideShare.
Whether your passion is information design, data engineering, statistics or something else entirely, the people you turn to for inspiration might be very different from your peers. So seek out and curate your own list of influencers based on your passion and interests.
Tip #2 - Never Stop Learning
The field of data analytics is comprised of a lot of different types of people who possess very different hard, soft and technical skill sets. Regardless of your area of expertise it's critical that you never stop stretching your mind and acquiring new knowledge. Today there’s no shortage of places to acquire new knowledge. Whether you go back to school full time, take a night course or take a free online course from Coursera, Udemy, Khan Academy, there so many amazing ways to be a life-long learner. The hard, however, is finding your focus so you can invest your time and brain power on acquiring the right skills.
Aside from going back to school, reading is of course a great way to keep stretching those cognitive muscles. And in this section of the presentation my essential reading list for analysts and information designers.
The full list is on SlideShare, but there's one book in particular that deserves a shout out, which is Edward Tufte's The Visual Display of Quantitative Information (VDQI). This is, honestly, the most important book for any information designer, or anyone broadly interested in how to effectively communicate with data. In many ways, VDQI did for the information design field what Graham & Dodd’s Security Analysis did for securities. The book is packed with fundamental lessons, ideas and concepts that are both timeless and essential to any field that deals with communicating with data.
Tip #3 - Seek Truth, Not Validation
For you to have any credibility in this field you need to ensure you're always seeking truth in the problem(s) you're trying to solve, and being truthful in how you communicate your findings/data to others. A while back I penned an article where I outlined 3 types of people who use data, which includes:
The Good (aka the truthful) - Those who use data to uncover truth regardless of the outcome, and who possess the skills to uncover it (e.g. analysis, critical thinking, etc)
The Bad (aka the deceptive) - Those who knowingly deceive their audience with data to push others to subscribe to their worldview or idea(s).
The Ugly (aka the ignorant) - Those who lack the knowledge and skills to apply or interpret data as a means to uncover truth. These people could have good intentions, but misinterpret or misuse data due to errors in their compiling, analysis or interpretation.
When it comes to The Bad (those who deceive), there’s one specific example I always like to share the rather infamous story of (former) U.S. senator Jason Chaffetz and his wildy deceptive use of data visualization. I penned a full analysis of the incident which you can read here.
The most important thing to remember here is that, as experts in this field we should always strive to be The Good (i.e. truth seekers), while calling out for The Bad (i.e. the deceivers), and teaching The Ugly (i.e. those who don't know any better).
Tip #4 - Always Strive to Make The Complex Simple
One of the challenging parts about this field is that many of business functions that it covers (be it engineering, data analysis, data science, etc) deal with relatively low level, technical and/or complex methods and concepts. And so we're often required to translate complex ideas, concepts or strategies to broader audiences.
Over the years I've found that some of the best analytics talent I've come across are people who can take something very complex and simplify it for broader consumption without losing the essence or substance of the concept. This task actually draws on what I think is a rather unique skill set that can be difficult to master - a capacity to recognize an individual or groups native business language and then be able to comfortably navigate both your own low level language and the audiences at the same time.
This skill reminds me of something I learned during my masters degree, which is the notion of abstraction levels in computer science. Simply put, levels of abstraction are used to present different models of "the same information and processes, but with varying amounts of detail". Lowers levels of abstraction are more technical and granular, while higher levels of abstraction tend to contain less detail and are more generalized. One of my professors often used the concept of abstraction levels to describe how different functions within a business need to communicate with each other by employing people who can traverse the various levels of abstraction, from higher business layers (i.e. C-suite) to low level production layers (i.e. developers, engineers, etc).
For me this concept perfectly describes what makes a truly great data analyst. Which is someone who possesses an ability to speak different business languages as a means to interpret ideas and simplify complex/technical concepts for different audiences.
Tip #5 - Be a Storyteller
Last but certainly not least is that you should always think of yourself as a story teller, and strive to tell engaging and compelling stories.
For me two of the most successful and gifted story tellers to ever were Charles Joseph Minard (French civil engineer famous for his data graphics) and Hans Rosling, founder of Gap Minder and renouned Ted Talks speaker.
So there you have it, my journey in the field of data and my tips for a fulfilling and lasting career. Hope you enjoyed it.