Get data science interview questions delivered to your inbox
Please enter a valid email address.

8 Data Science Projects You Can Do At Your Current Job

These data science projects can be done in any language, in any career.

Imagine if you had all the time in the world and you were also somehow magically funded. That would make it incredibly easy to get a data science job by practicing data science skills with side projects, at your leisure.

But most of us don’t live in this dreamworld.

When I was looking for a data science job, I read so many articles that told me to simply “do side projects” in my free time to get those tasty data science skills. The reality was that I had a full time job, a busy social life, a side hustle, and furry dependents (my cats). I can’t imagine how hard it is for folks who have actual human kids, or busier/more stressful jobs than mine. And to be totally honest, I was a little grudging about giving up my free time to do these data science projects.

My solution was simple: as a customer success rep, I found data science projects I could do as part of my current work. Most employers give you a bit of breathing room to pursue work tasks that help your job, but I did ask beforehand to make sure it was OK.

I scoped out a data science project - collecting data from marketing touchpoints to try to predict which resources were most successful in converting qualified leads. My boss was happy, the marketing team was happy, and I got to play around with some data that taught me data science skills.

No matter what your job is, you may be able to find a way to wedge in a data science task with your current work. This lets you save your free time, keep your job, and start building those skills that will help you get a data science job.

Data Science Projects to Consider at your Current Job

Here are the 8 data science projects you can do at your current job. These have been separated out into data science projects that make sense to do as part of a specific job, but depending on the flexibility of your boss and your personal interest, you could do any of these.

1. Sales: sales forecasting

If you’re on the sales team, there’s a lot you can do, so question yourself first to determine your own area of interest. Do you care about acquisition? Churn? Retention? Expansion?

Once you have an idea, put a question to it. For example, at one of my old jobs, I wondered whether I could increase expansion by scheduling more frequent, but shorter and informal calls. I could collect the data myself - my own calls, and the outcome. It was a long-term project, and ultimately I found that I didn’t increase expansions, but I did speed up the sales cycle with these shorter calls.

If you want to go deeper, here’s a list of five machine learning techniques for sales forecasting.

2. People Science: employee retention

People science is a growing field in many businesses. One of the biggest problems businesses face is attracting and retaining talent. For small businesses, this is especially critical to optimize since hiring is an area that doesn’t (immediately) generate revenue. Luckily, that makes it a great candidate for data science projects.

Employee retention is one of the best angles to explore. You’d probably have to be in this field to successfully do this project since many companies prefer not to share these details with all employees, but even if you’re not you can simulate this project with dummy data.

Here’s an example of a project you can do to calculate employee retention using data science.

3. Marketers: social media analysis

Marketing is a really interesting area for data science because there’s such a huge amount of data available. Most marketers are savvy with Google analytics and the various socials, so the challenge is to collate them and analyse them in a meaningful way rather than just spitting reports out of the Google dash.

It’s also an area that doesn’t require as much permission or difficulty acquiring data. Most social media platforms have an API that lets anyone grab read-access to basic data.

As a marketer, you can go in a lot of directions with this. Explore:

  • Which topics perform best for your competitors
  • Which channels bring the highest-quality leads
  • Where it makes the most sense to invest the budget

If you’re extra ambitious, you could even collaborate with sales to see which content brings in the most recurring revenue.

Here’s a beginner’s guide on social media network analysis.

4. Solutions Engineer: resource mapping

Solutions architects or engineers typically sit at the intersection of sales and technics. As they’re responsible for connecting customers with the support they need, there are plenty of data science project options to play with.

You could go in one direction and explore the technical cost of various projects or solutions that the team is building. You could also go the opposite and analyze which resources are best at converting customers.

Solutions engineers are the shepherds of some of the most valuable resources a company has at their disposal. Careful husbandry of those resources is a great data science project to undertake.

To be honest, most solutions engineers are doing data science day in and day out as Dennis Sawyers, a solutions architect for Microsoft, explains. It should be easy to find a way to discover what data science skills you’re missing, and build a data science project to address those skills.

5. Product Analytics: customer product use

Now that so many products are available online, it’s fairly easy for companies to perform product analytics: analyzing how products are being used by users and how features are performing. There’s a treasure trove of valuable insights to plumb right there.

If you’re in product analytics, or in an adjacent field, there are tons of angles you could explore:

  • What areas cause trouble for customers?
  • Are features being used as expected?
  • How can product use be streamlined?

This has implications for software engineers, technical support, sales, and even marketing potentially. Any insights you uncover would be valuable.

Here’s a list of examples of how to improve your product with customer use data.

6. Manager: predict employee performance

Managerial responsibilities are far-reaching. And unlike many of the other roles on this list, managers aren’t inherently technical. You may have to get creative to find data science projects you can do as a manager in your current role.

That said, there are definitely options. One potential data science project that overlaps with your existing job duties is predicting employee performance. This is helpful for managers who want to detect issues early and address them before they cause a real issue with the overall business.

All you need to start digging into this data science project is a comprehensive understanding of your KPIs. From there, you can map your employees’ performance against each KPI. This will help you build a predictive model to understand what are worrying factors, and what you can do to help employees under you in the future.

7. Technical Support: map out the ticket flow

For technical support staff, there are metrics that matter to ensure you’re within the SLAs. How quickly can you process tickets, and how high is the average satisfaction rating?

At its core, data science is just using statistical methods and analysis to answer a question. For anyone in technical support, or anyone who’s interested in customer satisfaction, one potential data science project you can do while still in your current job is mapping out the ticket flow.

Collect the data - the tickets, the customers, the topics, the outcome - and using statistical analysis and a visualization like a Sankey diagram, you can discover insights like:

  • What topics are most troublesome for customers?
  • Which customers take up the most resources?
  • Which agents offer the highest satisfaction rating?

Bonus points if you can automate the data collection and cleaning.

Here’s an example of a data science project walkthrough you could do at your current job.

8. Anyone: Clean data

It’s easy to think of data science as a glamorous job filled with dazzling graphs and important meetings. But the reality is that a lot of data science is finding a good way to get data and cleaning it into something usable. For anyone who hasn’t seen themselves on this list up till now, there’s one final data science project that works for anyone: build a data pipeline that can store clean data.

Here’s a great walkthrough of exactly how to build a data pipeline from scratch, as well as some pretty funny comics on the topic. To do this project, you’ll probably have to build an API and some automated data cleaning tasks. As for the subject, that’s entirely up to you. All it needs to do is answer a question.

This isn’t an exhaustive list of data science projects you can do at your job

By now, you should have the idea - find a problem or question in your field, find some data behind it, collect it, clean it, and interrogate it. It’s career-agnostic. All you need is interest, and the will to carry your project through.

And rest assured, this list is not the end of ideas. Data science projects you can do at your job are about as infinite as questions in your head. Curious about literally anything? Find the data, clean it, analyse it, and present it. That’s data science.

For more resources and inspiration, you can check out YouTube channels like Nate’s, which has a collection of data science projects you can draw insights from. Towards Data Science has a wealth of articles with walkthroughs and tutorials. Feel free to learn and modify any examples out there to learn data science.