The Case for Learning to Read, Speak, and Leverage the Language of Data
Why at least understanding data can be beneficial on a personal and professional front.
Learning how to work with data is just as (if not more) important as learning the basics of math, science, and other analytical skills. Not only is data analysis easier to learn than most STEM disciplines (because it is more task-oriented at its most basic level), but it takes a fraction of the time and is arguably more effective at making new learners feel like they’re walking away with actionable skills (most STEM education requires advanced to doctorate-level exposure to achieve confident mastery) to the point that they can begin demonstrating their understanding right away. It also serves as a foundation for learning more nuanced and advanced subjects, regardless of your level of exposure to those topics.
I don't know about you, but as a relatively new explorer in the world of data (just wrapped up my final term studying data analytics at Bachelor’s-level), I used to believe that learning how to work with data — specifically within languages like SQL and R — was only possible if it was something you picked up when you were younger. From my original point of view, it had to be an innate skill set built upon a natural inclination for analytical, primarily mathematical thinking. For that reason, I and others like me were likely intimidated by the notion of learning something with such a steep, seemingly esoteric learning curve.
As it turns out, my assumption was way off, as my imagination on this subject projected a false equivalency of anything data-related to the highest levels of data science (e.g. econometrics, advanced statistical programming, building algorithms, etc.). What I’ve found throughout my learning process, however, is that learning data analysis isn’t just about learning a series of tasks or using your skills and mathematical acumen to create a complex algorithm from scratch. To the contrary, the core understanding of how to work with data is actually founded in reframing how you think about a problem, which — in my opinion — is also the primary value of it to you.
What do I mean by that? Well, think about how people tend to approach problem-solving. In most cases, we want to get to C by manually iterating through A and B — usually in a fairly linear way. This takes extra time, and even with the investment of time, results in an inefficient result. In data analysis, those iterations can be linear, amorphous, or someplace in between without losing the informational infrastructure to codify your solution.
Here’s a case where this framework could apply:
You’re an SMB owner who wants to understand how what the average customer thinks about your small online business’s new product. You want to use this information to help you understand what they think of the product in its current state and you want it to inform version 2.0.
The linear path to this would be to use a run-of-the-mill online rating (e.g. Google, TrustPilot) system to provide some space for written feedback (usually, an open-ended survey). Then, you’d check in maybe once a week, month, or quarter to get a read on how people feel. This method will be fast and relatively easy to implement, often with low or no code needed aside from some light embedding that can be copied/pasted from a forum or using some hyperlinks in an email.
If you were to apply the data-centric frame of thinking, your conclusion would ideally start with a question rather than immediate action. Questions like:
- Do I want to understand more about my business or my product?
- In what form would I prefer this data to be presented?
- Is the point here to eliminate weaknesses in the way we communicate the product’s value or to improve my customers’ experience using it?
- Is the hope to use this information to restructure my team? My go-to-market strategy? My back office functions (operations, talent, etc.)?
Then, you would take those questions — for example, the one about using the data to improve a customer’s experience — and turn those into action items. These action items would consider the many ways to break each step of the process — from product inception to customer delivery — down to workable data points.
In the scenario where you want the feedback to inform product 2.0, you’d want to consider multiple elements for your data; a few potentially being engagement (where people click on your website, how long they visit it, etc.), UTM tracking (code that you can add to the end of a URL to track the performance of campaigns and content, per AgencyAnalytics) data, product (what product(s) is used most, usage rates versus total users on the site/platform, etc.) data, and GTM (lead or organic conversion rates, revenue per customer, cost to maintain per product, product churn [if applicable], etc.) data.
Once you’ve determined what elements matter most to the next iteration of your product and business, you can design a survey in such a way that these points are pre-defined, and thus, easier to aggregate when it comes time to develop actionable insights (e.g. dropdown/checkbox fields that pre-categorize customer feedback). The beauty of data-centric thinking is if you look hard enough, you can likely find an opportunity to automate this entire process for little to no cost — assuming you have no/a nominal amount of employees.
And this is just one use case for data skills.
Practical Rationale for This School of Thought
As demonstrated above, data analysis to me is about how to think about a problem; not so much about being “inclined” in any particular way. I’d even argue that grasping the basics is more tied to your ability to connect a question about a problem to its potential solution than it is about literal coding. Even on the coding side, mastering the fundamentals with practice supersedes smarts and talent.
The cool part here is even within the context of that aforementioned case, you can add or remove a slew of platforms, software, industries, skill sets, or other varied tools commensurate with your level of technical understanding while still applying your data skills, from Google Sheets to SQL or R — from MailChimp to Hubspot. The awesome thing is you’ll still get from A to C more efficiently.
I can provide specific examples until I have carpal tunnel syndrome and attempt to be persuasive until I’m blue in the face on this topic, but I’d prefer to allow the case to speak for itself.
That said, I will briefly explore some practical reasons why the value of [at the very least] learning how to think and assess a problem like a data professional is imperative in this day and age. These reasons transcend the need to use these skills in your day job, and instead demonstrate a more holistic value in applying this school of thought when facing just about any new experience, challenge, or question.
Reason 1: Data is ubiquitous and inevitable.
Data has always been everywhere. Sure, the applications that enable us to visualize, manipulate, and store it more effectively have improved (as have our tertiary economic use cases for it) tremendously, but it has existed since humankind recorded history and will only become more prudent with time.
An Example of This Reason in Action
Think about what something like a receipt really is. It could be defined as express consent to purchase the items you need in addition to providing a record of said transaction. From a pure data perspective, here’s what I see: A small, analog data set (that could be appended to previous receipts) of the brands, types of food, time of day, and money you’re willing to spend at a particular place.
Or, how about that paper report you receive after getting your vehicle serviced? In data terms, that isn’t just a report of the current state of your car. It could serve as a way to validate the projected performance of a new vehicle by a given auto manufacturer.
Let’s say Mazda expects oil life to last up to 7,500 miles in their 2024 Mazda 3. With an aggregated set of data from every service report across every Mazda Service Center, they could gather real data on the validity of that projection. Not just on the accuracy of that particular projection either, but Mazda could delve into even deeper insights like the influence of living in a more rural or suburban area (assumptions based on location data from the service center) on variables like the car’s oil life, brake integrity, and tire durability. And even for those who might say “Well I live in the city but prefer Service Center ____ out in the suburbs.”, the data can be weighted or structured in such a way that Mazda can still maintain confidence in their analysis.
If you train your eyes and ears to listen and look for it, data is everywhere — from your daily habits to what you like to do with your significant other. Once you can recognize those patterns in your daily life, it feels much easier to spot them at work, in school, or any other aspect of life that you hadn’t considered.
If everything in life can be translated into an accumulation of data points, would it not benefit us all to understand how to think in that language?
Reason 2: Learning how to work with data requires consistency, not innate intelligence.
As I previously mentioned, learning how to work with data in a more active manner (visualization, manipulation, cleaning, process optimization) is no different from learning to drive a car, lift a weight, or write a paper. It’s all about getting familiar enough with queues, mechanics, and vocabulary to feel confident in the solution you want to pursue.
An Example of This Reason in Action
What do I mean by this? Well, think about what it took to learn to drive a car. It wasn’t just about turning a steering wheel and hitting the gas where necessary; it was about preemptively assessing your environment for visual queues that inform how to turn the wheel. It was also about subconsciously learning to gauge the strength needed to come to a gradual stop or accelerate without ramming the car in front of you.
Of course, we never think of driving as a complex system of actions and concepts to grasp that it is because it feels familiar and socioeconomically urgent enough to bridge those gaps without overthinking. You would first learn how to think about data in the same way that you slowly learn each step of driving a car and navigating varied roads over time, starting with some vocabulary, concepts, and key terms. Then, you’d focus on how to perform basic functions that help you clean data (Note; Cleaning = structural uniformity for easier analysis), with no consideration for anything like a model or a strategy until you have a clear grasp of the basics. Then, you iterate on the basics each time you get into the car to drive.
Once you can understand that any data-related skill gaps or issues come down to how much you want to learn about it versus whether you can, it’ll become clear that learning it is no different than learning other skills built by repetition and intuition in our lives. Practice doesn’t make perfect, but it certainly builds proficiency, comfort, and discipline, which tends to have a compounding effect on achieving whatever level of mastery you’re seeking.
Reason 3: Building intuition about data will reduce your anxiety in the face of multilayered challenges.
A nihilist or fatalist might say that relegating all things to an amalgamation of data points proves the meaningless nature of life. A religious or more optimistic person might call it being pessimistic or denying a higher presence in our lives. I tend to explore it from a perspective that doesn’t aim to internalize this, however.
Data doesn’t dictate anything about us as individuals nor of our lives, but rather gives us basic insight into factual elements of reality in a given moment. The more you’re able to understand that “datatized” relationship between every little thing you do, see, taste, hear, and touch, the more you’ll understand that this is a mere fact of life — not a thing to judge or to feel judged by. If you can do that, I think you might feel less anxious about life.
So what does it mean when I say that thinking will reduce your anxiety? Well, in my purview, if you can look at the world through that lens, it unlocks another dimension through which to perceive everything that happens around you. The important thing here though is to remember this isn’t all there is to the picture.
An Example of This Reason in Action
Let’s use a universally applicable example: Social media. I expounded on this concept in another article, but let’s say that you’re caught in a doom-scrolling cycle. This cycle effectively serves as the most well-crafted feedback loop and self-fulfilling prophecy one could conceive of, where the longer and more intently you scroll, the more you will see the things exacerbating said scrolling, which consciously and subconsciously reinforces the impulse to scroll in the first place.
We tend to hate when we feel bad about things, but the line separating familiarity and comfort is razor thin, so it takes no time to grow so accustomed to this habit that even when we recognize the negative outputs, we fail to break the cycle of inputs. This is partially due to many chemical reactions that influence our brain and hormone activity — concepts way over my head. The consequences, however, are clear as day, with anxiety being one of the major symptoms in those over-invested in the world behind their screens.
Now, let’s imagine how differently this might go for those who learn to view things like doom-scrolling through a lens of data analysis (while keeping our humanity, of course). When you see a negative thing, rather than immediately succumb to your emotional reaction, you might ask these questions instead:
- “Why am I being shown this? Why is it being delivered in this way?”
- “Is this for my betterment or is it merely a piece of content from a long list of content that is being prioritized to me based on previous things I’ve watched?”
- “If so, is the content even the point of my being shown, or is this merely a numbers game, assuming advertisements are also being run on this feed and someone stands to benefit from my engagement?”
- “What are they trying to sell to me here? An idea? A product? A service?”
- “If the answer to any of these questions leans toward yes or has an antithetical answer, why am I allowing this to disrupt my day?”
This is just one example, and it might appear to be overthought, but as a person with some tech work experience, I can assure you with reasonable confidence that this is a common strategy from the other end because engagement and retention are paramount to the success of any product or company, even if the only point of the engagement is to garner more engagement to appease advertisers with deep pockets.
This methodology is no different from upselling a premium product except that the venue is the palm of your hand rather than a car dealership or furniture store, and the frequency is 1,000x+ the rate at which you’d receive that sales proposal in a physical space. It also isn’t innately evil per se, as there isn’t a specific person with their finger hovering over a big red button ready to give you 20 more reasons to hate or fear Topic A or B, but it is the mark of an effective algorithm, which isn’t always great for our mental health.
Once you get to the point where you can deconstruct most things that you see, from political ads on TV to inflammatory videos on Facebook, you’ll view them as the compilation of data-driven selling points that they often are. This framework transcends hot-button issues as well.
Reason 4: Learning data analysis will make it easier to learn even harder concepts.
Circling back to one of my earlier statements —
“It also serves as a foundation for learning more nuanced and advanced subjects, regardless of your level of exposure to those topics.”
— I’ve never in my life felt more mental acuity than when I began to develop a data-centric frame of mind, and it had nothing to do with R, Excel, or any of the more technical knowledge. The benefit was much more straightforward.
An Example of This Reason in Action
Think about the complexity of learning something like the impact of climate change on our oceans, for example (example comes directly from my Oceanography class at UPenn). This area of science requires at least a loose grasp of biology, chemistry, and physics — not to mention the more nuanced sociopolitical factors that contribute to this issue (e.g., debates around the fishing industry, tourism, shoreline dwelling, natural resource collection, etc.) if you want to understand it holistically.
Without a data analysis framework, you might try to apply the typical memorization model, focusing on vocabulary and key topics within the larger subject that pertains to the specific problem you want to understand. There’s technically nothing wrong with this approach, but let’s consider three major blockers here:
- Do you have enough insight to understand what context is missing from what you’re learning?
- Could you, just from vocabulary and reading, cite meaningful numbers that support your statements?
- Can you come to a reasonable, succinct conclusion that you could talk through with someone else without hours or days of prep?
The chance that your answer to any of these questions won’t work in your favor is fairly high. From a data-oriented framework, you can glean understanding through the application of the skills and perspective data analysis innately offers you through your analysis. Understanding how to intertwine data collected between historic ocean temperatures, surface winds, and natural disasters doesn’t take climate expertise. It takes an understanding of how to combine data to extract more meaningful insights. And while you can develop a perspective on climate issues and data over time as you seek ways to validate your analysis, it won’t require a doctoral degree.
If this is true under these circumstances, imagine the scale of challenges you could take on that you might have previously written yourself off for!
Reason 5: Achieving Generalized Specialization
General specialties are essentially the skills that transfer easily from one job to the next. Within the context of my reason for including it here, data skills — though specialized in practice — are generally applicable wherever you go because strategy, substance, and focal points may change, but the fundamentals required to conduct a sound analysis will remain the same.
Those with data skills already understand this, but if you were ever exploring the idea of learning a data-centric coding language or becoming a spreadsheet wiz for your nonprofit organization, the concept of generalized specialization essentially confirms that learning how to work with data in a meaningful way is never a waste of time.
You don’t have to be an aspiring data scientist or AI researcher to derive value from learning data skills, and the cool part about becoming a general specialty is if you learn to contextualize what you already know how to do in a way that others haven’t, the sky becomes the limit.
At the end of the day, learning how to work with data is what you make of it. Since we aren’t limited by the parameters of traditional schooling, there are no negative or positive implications to what you choose to pursue. I’ve just spent quite a bit of time learning the methodologies and frameworks underlying data analysis over the past couple of years, and I have already received a great deal of value from that undertaking.
Just as I’ve described throughout this selection, data and its role in your life is inevitable. Almost every element of your work and life has been “datatized” to some extent, if not to the fullest extent. So, whether to be a practitioner or merely to understand the current state of life as we know it, I am a strong proponent of incorporating data literacy and skills as life skills, because they have arguably become just as important as the subjects we’re taught from our first years in school to high school graduation and college. Data skills are not nice to have — they are fundamental to our existence in an increasingly digitized and automated world.
Life won’t end without them, but it could be harder. And you certainly don’t need them to succeed in this day and age, per se, but it could reduce a great deal of friction in an effective, simple way for whatever you do. I certainly am nothing close to a genius, but learning these skills has been transformative for me and the way I’ve chosen to engage the world around me. I think the demystification of this topic could drive exponential value for others like me who might’ve undersold their ability to learn something new.