Lately, a lot of articles making the rounds on the internet are talking about data science as one of the best jobs you can aspire to. There’s plenty of reasons for that. There are roles available in seemingly any company and field. The pay is great. You can tackle a lot of complex challenges that need creativity and wit. The data science community is thriving and exciting.
Yet, for all the good things data science has to it, there are also a couple of things that aren’t that glamorous. The worst part is that it seems practically no one is talking about them! So, it’s only fair for any aspiring data scientist to get the full picture and not be deceived by the siren song crafted around the field. There are a lot of things to love about data science but you should only aspire to work in the field if you know the following 4 things.
1. Day-to-day Work can be a Challenge in Itself
As we said above, one of the exciting things about data science is that it’s a field that comes packed with challenges. Working with complex algorithms to derive insights from huge data sets certainly feels like an essential task for businesses and organizations across a wide diversity of fields. No wonder why so many aspiring data scientists believe they’ll end up dealing with sophisticated solutions to develop new medicines, foster transport innovations, or drive major growth.
However, the reality is somewhat far from all that, especially for junior data scientists. Rather than working on thrilling challenges on a daily basis, you’ll find out that the day-to-day work can be a challenge in itself. That’s because you’ll have to deal with things that don’t feel as transcendental as the ones cited above.
Case in point – you’ll have to handle the data science infrastructure or develop a strategy from the ground up. Though that feels like a no brainer, you’d be surprised by how many companies don’t have a data plan in place before they hire a data scientist. If that happens to be you, you’ll have to deal with the business expectations and with the lack of a proper context to develop a sound strategy.
Additionally, you might expect to work in a disruptive data-driven solution only to find yourself working with a company that only wants a small part of data science. In other words, not everyone out there is trying to use data to create a revolutionary experience, so chances are you’ll be working for a company that just wants enough to make safe business decisions. That’s all fine and dandy but if you’ve come to expect more from your role, you’ll have to face the challenges that reality itself will pose to you.
You can hear more about what’s being a modern data scientist in this great TEDTalk by Asitang Mishra, a data scientist at the NASA Jet Propulsion Laboratory:
2. You Might End Up Being a One-Man Team
Here’s another harsh reality of the data science field – a lot of companies aren’t giving it the relevance you may expect for such a potentially disruptive discipline. That means that you can be hired by a business that will mostly depend on you and you alone to move the data science strategy forward. Unfortunately, this translates into two things.
First, you’ll have to create value almost on your own, which is very hard to do. That’s because data science doesn’t exist in a vacuum. To truly benefit from a data science team, the company has to ensure that said team is in touch with the rest of the teams. Only by collaborating with other teams and experts will data science provide the solutions that can take a business into a winning strategy.
Second, by isolating you and/or the data science team, the company will put an insane amount of pressure on you. That’s a recipe for disaster. As you’ll be expected to deliver value quickly and with virtually no help, you’ll face a lot of issues and frustrations along the way. In the end, you’ll be mostly set up for failure, which can put you and your career as a data scientist under stress.
Of course, this doesn’t happen in all companies, but since a lot of business leaders still don’t quite grasp the capital importance of having a data strategy, it’s better if you’re warned about this.
3. Everything Remotely Related to Data Will Come to You
This is a tough one. When you become a data scientist, a lot of people start assuming that anything that’s somewhat close to data is within your purview. And that’s an issue because, even if you do know a lot of stuff about data methods, techniques, and strategies, you can’t possibly know everything.
For instance, some execs believe that they can task data scientists with analytics reports about web performance while also looking for their assistance with reports about sales or database upgrades. Sure, these all have data. But that doesn’t mean you have to know everything about them!
What’s more – some businesses expect that since you handle data, you have to provide answers to anything and everything they can think of. Need a new customer segmentation? Trying to optimize the sales process? Searching for ways to limit bottlenecks in distributions? Ask the data guy!
Unfortunately, that’s not how data science works. In reality, you’ll have to create tools to help the people in marketing, sales, distributions, and so on make their own decisions applying their knowledge about their subjects. So, trying to make everyone understand that you don’t actually have all the answers becomes an additional chore for data scientists.
4. You Shouldn’t Count Data Science Competitions as Experience
Finally, there’s a misconception that’s been around for quite some time that aspiring data scientists need to know – seeing competitions as experience. Look, that doesn’t mean that data science competitions aren’t beneficial. They have become so common in the last few years because they are actually good to build skills, network, and even land a job.
But from a professional standpoint, you shouldn’t consider them as experience. The main reason why is that the projects you’ll find in the real world are entirely different from those you see in competitions. The datasets you’ll handle in those contests are doctored to look spotless and you’re mostly evaluated for your accuracy.
However, real-life projects are always messier and more chaotic. Datasets are mostly comprised of information coming from multiple sources that often needs to be collected and organized. In fact, you’ll spend a lot of your data science job trying to tidy up those datasets before feeding them into an algorithm.
Additionally, all your data science solutions will have to work in a bigger context. That means that your algorithms will have to serve different purposes while remaining as simple as possible. Since interpretability is key in corporate scenarios, you’ll have to worry more about the effectiveness of your solutions rather than on increasing the accuracy. This, of course, is a radical departure from what you see in competitions, so the experience you get in these is unlikely to apply.
Bursting the Bubble
All of this can sound discouraging to a lot of people aspiring to be data scientists. If you’re one of them and are feeling like that after reading this article, please don’t. The idea isn’t to drive away people from the complex and challenging world of data science. Rather, the objective is to provide a clearer picture of what the field is about.
As it happens with a lot of jobs (especially emerging ones that deal with cutting-edge technologies), the role of the data scientists is still misunderstood. That’s why it is important to shed some light on what to expect and what the job looks like today to avoid frustrations. As you can see, some of the issues data scientists have to face lie mainly in companies failing to see beyond the promises.
Fighting against that is only possible by creating reasonable expectations, getting rid of the unnecessary glamour surrounding the field, and comprehending that it’s a daring and relentless sector. However, as far as complex fields go, you’d be hard-pressed to find one as defining and challenging one as data science. So, if you’re the kind of person that enjoys that, then go ahead with data science – but keep the 4 things above in mind.