Recently a friend asked for advice on what to look for when hiring a quantitative person for their impact investing organization. Here are the qualifications I recommend:
Empathy Grounded in Experience
An ideal candidate will have experience both as a producer and consumer of data. In other words, they should understand what it’s like to request data from busy entrepreneurs and also what it’s like to be the busy entrepreneur.
Evaluation Strategy: There’s no surefire strategy for determining who is genuinely empathetic, but the right interview questions can go a long way. I recommend the following:
- Tell me about the last time you requested a large amount of data from someone. How did you ask? Were they able to provide the data on time? (If not, why not?)
- Tell me about a time someone asked you for data you couldn’t provide. How did you respond?
Editorial Mindset
To be an effective impact investing quant, you need to think like a good editor. This means asking the right questions, quickly understanding how disparate facts connect to form the big picture, and eliminating distractions to get at what matters.
Evaluation Strategy: Ask the applicant to design a short survey to evaluate a specified outcome. Once they’ve written the survey, discuss the logic behind their question choices. The goal here is to see if they can take an abstract problem and drill down to clear and reasonable questions.
Quantitative Analysis and Presentation Skills
Surprise! An impact investing quant needs to have strong data analysis and presentation skills. There’s no single profile but at a minimum the applicants should have the following:
- Ability to tidy raw data from a wide variety of sources quickly and without error. This means everything from poorly formatted spreadsheets, to unstructured text, to vendor APIs. The more data an analyst can access quickly, the larger the universe of potential solutions will be.
- Demonstrated ability using statistics for exploration, inference, prediction, and optimization.
- Deep understanding of statistical pitfalls and remedies.
- Ability to create new measures or metrics from scratch.
- Demonstrated ability to present complex quantitative research to general audiences in writing and through data visualization.
- Strong Excel/Google Sheets skills and competency using another common data analysis tool. I think R is ideal because it’s free, incredibly flexible, and has a huge (and helpful) user base. Other useful tools include SQL, Python, Tableau SAS, Stata or SPSS depending on what your organization uses.
Evaluation Strategy: To evaluate a candidate’s data skillset quickly, I like to ask the following questions:
- What was the last quantitative technique you learned? How have you applied this technique in your work?
- What was the last data analysis or visualization tool you learned to use? How have you used that tool in your work?
- In your previous work, did you ever develop a new way to measure something important? If so, how did you develop the measure and was it effective?
If they make it past the interview, there’s no substitute for a real-world task. Send the applicant a sample dataset similar to one they’d encounter in your organization and ask them to produce a short report with graphics in a limited amount of time.