From Slack Huddles to Code Reviews

Are you fascinated by the world of marketing and analytics? Do you enjoy analyzing data and resolving business challenges? As a lead marketing data scientist for an e-commerce group, I am fortunate to engage in these activities daily. I work alongside a small yet powerful team to manage various descriptive and predictive analytic projects that fall under the marketing and consumer analytics spectrum of the business. In this blog, I will guide you through a regular day in my position, from morning meetings to code evaluations, and everything in between. So if you’re curious about the responsibilities of a marketing data scientist or if you’re considering a career in this field, read on!

One of the first things I do every morning is hold a quick Slack huddle with my team. We discuss our goals for the day and any technical or functional challenges we may be facing. Technical challenges could be data feed issues from vendors or other departments, while functional challenges could be things like unclear project requirements from internal business stakeholders.

Brainstorming solutions with an open mind in the morning is the best team management exercise I’ve done so far. It helps us start the day on a positive note, knowing we have a team to rely on and that we’re all working together toward our goals. At the end of the day, we can look back and see how far we’ve come, and that’s an incredible feeling.

We typically work on two types of projects every day: descriptive and predictive analytics. Descriptive projects focus on automating marketing analytics reporting frameworks that provide point-in-time data such as year-on-year, month-on-month, week-on-week, day-on-day, and even hour-on-hour sales, revenue, inventory, and profit data from our in-house ERP systems. This data can be dissected by category, sub-category, brand, SKU, state, city, and more, and is available on a near-real-time scale. Orchestrating these cloud feeds in near-real-time is like providing business stakeholders with a treasure map to navigate the vast sales and marketing universe.

On the other hand, the predictive modeling projects we handle are more about using our data to figure out what’s going to happen next. For example, we use machine learning to group our customers by how valuable they are (Customer segmentation & Lifetime value) or to predict whether they’re likely to leave us or buy more from us (Propensity to churn & repurchase). We also use our data sources to personalize product recommendations for each customer (Customer Recommendation Engines) or to see which digital marketing channels are working best for us (heuristic and algorithmic attribution modeling). Also, we predict forthcoming data points (forecasting) and help the business understand what customers truly appreciate or criticize the most by doing audience text mining on customer reviews, NPS surveys, chat logs, and more (Natural Language Processing). It’s always fascinating to see what customers have to say and to find ways to improve their experience. After all, happy customers mean a thriving business!

Anyway, after our morning Slack huddles, I usually have catch-up meetings with managers and stakeholders from different business units, such as customer support and sales or say, logistics, delivery and supply chain to business strategy, and market retention teams - it really depends on the day of the week. During these meetings, we talk about their side of the business or work together to solve a business problem using our existing marketing data science and analytics cloud frameworks. I always look forward to collaborating with other teams to get a better understanding of the business and see how we can use data to create a positive impact on customer experience.

If you’re new to the senior data science and analytics role, let me give you a piece of advice. Trust me, it’s a great idea to make simple and straightforward communication with other departments a daily habit. As you grow into this role, you’ll realize that technical intricacies can sometimes cause confusion and make it difficult to get your message across. So, don’t be afraid to use plain language and take the time to explain your ideas to your colleagues who consume your business intelligence models or API frameworks. By doing so, you’ll build stronger relationships and create a culture of collaboration that will help you achieve your goals more effectively.

Finally, every week, I get to sit down with the big guns - the Head of Marketing and other senior leaders - to figure out which projects we need to prioritize next. It’s awesome to see how all of our hard work fits into the company’s overall goals, and it gives us a chance to showcase our progress. Plus, it’s a great opportunity to improve our processes and make sure we’re always delivering the best possible results.

So, I think that’s one side of my work; anything and everything around managing internal clients, teams, and vendors. It’s all about keeping everyone in sync and making sure things run smoothly. But that’s just one piece of the puzzle. The other side of my job is where I really get to flex my data science skills.

Every day, I try to set aside 1 to 3 hours to work on code snippets, data visualizations, and code reviews. It’s a collaborative effort with my team, and we use a variety of tools like SQL, R, Python, and Google Cloud platforms like Looker, Vertex AI Jupyter Notebooks, Cloud Storage Buckets, and Big Query. I have to say, this is one of my favorite parts of the job. Not only do I get to work closely with my team, but I also get to learn so much from them. It’s a win-win situation.

If I don’t document everything that we work on throughout the day, it feels like something is missing. So, every day, I make sure to spend quality time documenting our work on Trello Boards, Git versioning systems (CI/CD pipelines), and Confluence pages. This not only helps us keep track of what we have accomplished, but it also helps us communicate and share our progress with internal stakeholders and team members. Plus, it makes it easier to pick up where we left off the next day, and it’s always satisfying to see how much we’ve accomplished over time.

So yeah, I think my role as a lead marketing data scientist at the e-commerce group is one that requires a diverse skillset, a passion for problem-solving, and a love for analytics. It is a career path that I find incredibly rewarding and exciting, as I get to work with a talented team of individuals, collaborate with various business units, and use cutting-edge technology to drive business growth and success.

If you’re a newbie interested in marketing and analytics, I encourage you to explore this path and see where it takes you. With the right mindset, determination, and willingness to learn, you could be well on your way to a fulfilling and rewarding career as a marketing data scientist.

Congratulations if you have made it this far! I think that’s pretty much everything I would like to convey for now. I hope this real-world example has given you a behind-the-scenes look at E-commerce from the perspective of a marketing data scientist. I’ll write to you soon.

Thank you.