Many companies realize they cannot fulfill their Big Data dreams without finding the right talent, so they’re hoping they can somehow lure these elusive, valuable resources into their organization. My advice: Be careful what you hope for, you just might get it. Of course, all the fancy technology in the world does no good without skilled talent to analyze the data. However, acquiring talent is much different than procuring technology. For better or worse, talent comes with culture, and that culture can make or break your company. Here is my advice for acquiring and integrating data scientists into your organization.
Finding great talent
Contrary to popular belief, good resources for Big Data are in abundant supply and that trend will continue to increase–perhaps dramatically–over the next five years. There is a global awareness of the need for people who perform advanced analyses and bright people are taking notice. Many of the top universities are offering advanced degree programs specifically structured for data analysis. I’m also noticing the emergence of boutique recruiting firms who specialize in analytic talent.
Regardless of the new Big Data trend, data analysts have always been easy to find if you know where to look. Data analysis is not a new discipline–it’s not even an emerging discipline. Bayesian theory has been around since the early 18th century and we’ve seen statistics and advanced data analysis applied to business and financial concerns for many decades. Mathematics is a fundamental science, and although not everyone has the talent and inclination to study advanced forms of mathematics, some of us do. I took trigonometry, math analysis, and advanced placement calculus in high school and my classes were filled with other teenagers who were very bright.
People are concerned about the engineering and science talent that’s emerging from US colleges today–and they should be; however, according to the US census, we still put out about eight to ninety thousand engineers per year. That’s several hundred thousand engineers on the market just in the last few years in the US alone. We haven’t begun to talk about all the engineering talent emerging from around the world including India, China, and Russia.
Trust me, there’s no shortage of data analysts–there’s a shortage of proper expectations. Everybody has Big Data dreams, but nobody wants to pay Big Data prices for these resources. Companies need to get real if they plan to actually mobilize their Big Data strategy. A good data scientist can easily make $300 – $350K per year. Quants (Google it) can make $500K+ under the right conditions. So don’t get frustrated when you cannot attract a quality data scientist for $120K plus free movie tickets every quarter. Also, the idea that you’re going to find a data scientist that can also handle project management, executive communication, business requirements, operations, software development, and the occasional general journal reconciliation–is just insane. The job design is simple; data scientists analyze data.
To succeed with data scientists you must understand the culture they bring with them. If you ignore this, you stand the risk of a culture clash you’ll never forget. There’s no sense in spending a bundle on a team of data scientists, only to deal with their disruption upon arrival. All data scientists come with their own culture which is attached to their profession. I’ll paint the culture of the profession with broad strokes; for actual integration, you’ll want to consult a specialist like myself who understands both the talent and the culture. It’s also important to consider the national culture of your data scientists, especially if there’s a dominant representation from one nationality within your group. As you can imagine, a team of data scientists from Chennai will be culturally distinct from a team of data scientists from Boston. Notwithstanding national differences, there are cultural generalities worth discussing that apply to most data scientists.
The most notable characteristic of data scientists is their love for excellence–sometimes to a fault. They have excelled in school their entire life and they pride themselves on their acute understanding of their realm. This confidence often extends fallaciously into subjects they know little about. For instance, don’t count on them for leadership–that’s why you have leaders. Their quest for excellence can also put your innovation at risk. They would rather see an excellent product delivered late than a sufficient product delivered on time. Never rely on a data scientist’s advice on whether an information product is ready to go to market–that’s why you have product managers.
Data scientists rely on science for solutions and generally view human involvement as a huge source of error. When I studied to become a Six Sigma Black Belt at Motorola, one of my Master Black Belts told me Motorola was trying to improve Six Sigma by eliminating as much human interaction as possible. Although the complete absence of human involvement lives in their world–in the real world, people matter. Sometimes a leader’s decision contradicts what the data would suggest. This can be a point of contention with data scientists that needs to be managed.
Finally, data scientists are usually introverts that like what’s going on in their own head more than they like what’s going on in the outside world. They can be social (especially the Silicon Valley variety); however, socialization drains their energy quickly. To recharge, they need quiet time with themselves. The behavior attached to their most productive time is typically confused with unproductive behavior: like staring out a window, or playing a video game. When under too much stress, this behavior is exacerbated. They typically withdraw from society and refuse to engage with anyone. You must develop strategies for recognizing this behavior and reducing stress levels before things get out of control. They may be brilliant, but they’re also human.
If you’re looking to reel in the Big Data fish, make sure you have a big enough boat. Big Data talent is all around, but you must be in touch with the real costs of acquiring these resources if you ever expect to move your Big Data strategy forward. It’s best to talk with industry experts who are familiar with the market’s expectations or conduct your own surveys before launching an acquisition campaign. Once you have them in house, expect that they will seek excellence at all cost, exert their expert power everywhere, complain when their analysis is ignored or overruled, and clam up when things get stressful. And once in a while, they’ll discover a breakthrough innovation that feeds everyone in the organization for five to ten years. It’s a worthwhile pursuit; I just want to make sure you know what you’re signing up for.