Do you hire like a hunter or a farmer?
There are two philosophies for growing companies. Farmers focus on hiring inexperienced people and cultivate them to be effective. Hunters look for trophies (stars, “the best”) and expect them to be effective from day 1. Carol Dweck would say hunters have a fixed mindset while farmers have a growth mindset. Hunter culture can be short-term effective, but it is usually not sustainable long term. In fast-paced, competitive business, hunting is seen as superior to staid, complacent farming. This is silly. Farmer culture embraces diversity, collaboration, and employee growth. Both hunting and farming require patience. With an emphasis on teaching and employee growth, farming can have greater impact and be more cost effective than the hunter approach.
Historically, companies in technology and investment banking (the two industries where I’ve worked) have hunter mentality, looking for world-class talent. Netflix is a hunter: “Like all great companies, we strive to hire the best.” Amazon is a hunter. Hedge funds are typically hunters, focusing on hiring from “top” schools and emphasizing the number of PhDs in their midst. Laslo Bock showed that common hunter metrics, like GPA and alma mater have no predictive power when it comes to employee success after hiring. They also fall prey to structural biases related to privilege.
Hunter culture tends to replace “non-performers” quickly. If someone is not performing well, then that implies they aren’t the best, so they’re out. This can foster cutthroat competition and paranoia, not to mention burnout. Another consequence is that there’s a finite number of “superstars”, so hiring is very difficult, time consuming, with a delayed payout.
The language of diversity and inclusion is on the rise, but there is still resistance to farmer culture. One common argument against is that it takes too long. This is a classic trade-off between time versus money. While inexperienced people are less expensive than experienced people, it takes time to train them. Is it worth training people or is it better to hire people with specific skills you need?
One way to answer this question is to model employee experience and employee impact via a Monte Carlo simulation. Simulation is considered the third branch of science and is an inexpensive way to understand the behavior of a population or system. For my startup, I model employees as having one of four shapes. This is inspired by Valve’s ideal T-shaped employee. What other employee shapes are there? Valve doesn’t say, because they only hire Ts. Here’s how we think about it at Pez.AI. Most university grads are dots: they know a little bit about something. Next are specialists. People with graduate degrees and work experience in one area tend to be specialists (I-shaped) because they know a lot about something. Then comes the T-shaped employees that know a bit about many things and a lot about something. I refer to these people as specialized generalists. Finally, there are masters. They know a lot about many things, which visually looks like a filled in U (or D).
Hiring only T-shaped employees can lead to many wonders, as Valve has shown. But it can lead to elitism, which is the opposite of inclusion. As a former professor, I think it’s better to hire dots and cultivate them to T or beyond. Some people may stay a novice for a while, while others quickly move to T, depending on their background, motivation, mentoring. The likelihood of moving from one shape to another is known as a transition probability. Combined, these probabilities form a transition matrix, which defines a Markov chain. In my simple model, most people stay in the same state from one quarter to the next. So 80% of novices will remain a novice in the next quarter, while 10% will acquire enough expertise in a skill that they become a specialist.
What about those going “backwards” from specialist to novice? This can be explained by people making lateral moves and “forgetting” their previous experience. It can also represent people who are so underutilized that they are effectively novices.
One of the interesting things about a Markov chain is that no matter the starting configuration, the final distribution of states is the same. In other words, it’s guaranteed to converge to a particular distribution, given enough iterations. In the below charts, I’ve simulated 10,000 employees hired as either all novices or all specialists. The numbers at the top of each stack of bars represent a dollar cost (in thousands), based on 80k for a novice, 120k for a specialist, 250k for a specialized generalist, and 400k for a master. These numbers are reasonable in NYC in either the tech or investment banking world. In the first year or so, the farmer approach (hiring inexperienced staff) is about 33% cheaper than the hunter approach (hiring experienced staff).
After about 3 years (12 iterations), the probabilities have stabilized regardless of the initial skill states. Why is this significant? Assuming that your company culture and training programs determine the transition matrix, the skill state distributions will eventually settle into its own equilibrium. Put another way, your culture is more important than who you hire. The farmer approach recognizes and embraces this, hence focusing on cultivating top talent, rather than searching for top talent. With a strong commitment to growing your employees, it’s possible to speed up the transition from low impact states to high impact states. This changes the transition matrix, resulting in a more optimal configuration. Sure, you still need to select for people that you think will succeed in the culture, but more effort can be spent cultivating than searching (which has both opportunity cost and monetary cost). In a culture of continuous learning, employee development provides a recurring dividend of increased productivity and sustainability.
So which are you? Do you have an effective farmer or hunter culture? Share your experience (and data) in the comments. Or just give us a clap!
Brian Lee Yung Rowe is Founder of Pez.AI, the chatbot platform for data-driven digital empowerment. Join the beta to see how bots can increase the impact of your organization.
