Meet Halla, an expert data scientist based in Chicago. Born in Germany and raised in Indiana and Tennessee, Halla holds a Ph.D. in Business Economics from Harvard Business School and a B.A. in Physics, summa cum laude, from Harvard University. Prior to McKinsey, Halla developed trading strategies using natural language processing at Jump Trading; served as a senior portfolio manager at Arrowstreet Capital; and managed U.S. equity portfolios at Goldman Sachs. In his free time, he competes in forecasting competitions and has finished in the top ten four times against hundreds of competitors, including a 1st place finish in predicting prescription drug volume for Pfizer Oncology (learn more). He and his wife Liz (a theatre professor at Northwestern University) have two children, Sofia (age 6) and Isaac (age 4). Here’s a peek into Halla’s day-to-day:
7:00am: Look at results from analysis I ran overnight; see if they make sense and formulate any implications I see from the data. I also respond to any emails that have come in overnight from overseas.
8:00am: Grab coffee and breakfast and head to client site. My commute usually takes about an hour door to door. When I arrive, I check in with my team.
9:30am: Problem solving session with my McKinsey team on the customer clustering portion of our analysis, trying to better understand which segments of customers are likely to become loyal and which are likely to churn.
10:30am: Meet with our client to review the latest sales forecasting model by channel. The client says channel A should be a priority because they cost less to service, and would like to see some models specific to this segment.
12:00pm: Lunch with team. Today we go to a new Mexican restaurant right around the corner from the client site. We try to use lunch time to get to know each other better outside of work so we talk about our weekends. The engagement manager on the team shares his recipe for beating eggs with leeks and caramelized onions, while a team member from Denmark praises the ease of sous vide cooking. I share my love for grilling meat in the back yard, even when there’s snow on the ground.
1:00pm: Call with another team working on a different problem in the chemical manufacturing space. Based on my experience, I’m able to suggest some new ways to think about selecting the best set of features for forecasting the production efficiency of their smelting process.
2:00pm: Write more code, run more analyses.
3:00pm: Another problem solving session on customer segmentation, this time with a marketing analytics expert who flew in from San Francisco.
4:00pm: I take a step back from the analyses and team discussions to capture some of our findings and recommendations on PowerPoint slides. I draft them on paper by hand, sketching out how I want them to look. Then I send them, along with some data sets to our visual graphics team who will create them and clean them up overnight (it’s almost magical).
5:30pm: Check out with my team and head back to the hotel for dinner.
8:00pm: After a break, some food, and a call home to my family, I usually do a bit more exploratory data analysis. I experiment with different ways to visualize the data to dig into unusual patterns, and find that certain business lines appear to have experienced a simultaneous productivity spike in August of last year. I make a note to follow-up with the client about this and ask questions that may help us improve our models.
9:00pm: Read about projects from other teams within the broader analytics community at McKinsey. I love working on such a global team and some of our internal communications give me ideas for things to try on my projects and introduce me to people I can connect with to expand my network at the firm.
10:00pm: Conduct a phone interview with an analytics candidate for an office in India. Helping to build the firm and our practice by working with potential new members gives me energy.
11:00pm: Lights out.