Data science, the Internet of Things and big data are terms we hear a lot these days, but did you know that the big data revolution is a recent phenomenon? The term data scientist has been around for less than ten years. Now every company that desires to improve its bottom line wants data scientists and to understand big data.
Data Scientists are scarce and in high demand, so the laws of supply and demand dictate that they command extremely high salaries. The August 2018 LinkedIn Workforce Report shows a shortage of data scientists, with the situation acute in many cities such as New York or San Francisco, and even in smaller cities such as Austin. A search on Glassdoor for “data scientist” shows that the base salary for a data scientist just starting is around $101,000 a year. The average base pay is about $140,000, and this can rise to close to $200,000 a year. This means that only the largest, most established organizations can afford to hire data scientists.
Whereas in the past data scientists worked mostly in the high-tech and finance sectors, today every industry is collecting data and wants to interpret it. But, if an organization collects data and then doesn’t use it, it becomes stale and can’t influence decision making. What makes big data useful is not purely the fact that it is information that can be studied to gain valuable insights on consumers; it's also the role it plays in developing effective machine learning — and, by extension, deep learning — neural networks and algorithms.
An article in the Harvard Business Review (You Don’t Have to Be a Data Scientist to Fill This Must-Have Analytics Role) agrees that “data scientists are required to build the analytics models — including machine learning and, increasingly, deep learning — capable of turning vast amounts of data into insights.” But it goes on to say that “More recently, however, companies have widened their aperture, recognizing that success with AI and analytics requires not just data scientists but entire cross-functional, agile teams that include data engineers, data architects, data-visualization experts, and — perhaps most important — [analytic] translators.” That's a long list of people to add to your organization to be successful with your homegrown AI and analytical initiatives, and it gives a sense of the scope of investment required.
The idea behind using data to accurately predict what your customers are going to want in the future can provide any organization with substantial growth opportunities. The problem, however, is that most executives don’t know where to start when it comes to implementing an analytics-based strategy. They need to understand what they want to achieve by hiring a data scientist and what predictions they require from their analytics.
They know that using predictive analytics should make them more money down the road, but they underestimate the time and effort required to mine their data meaningfully. In an article by Joseph Schott (How to Hire Data Scientists Based on Your Company Readiness published in datascience.com), says that when to hire a data scientist depends a lot on the specific stage of data maturity and adoption of analytics of a company. He has defined three stages, the crawl, walk, and run stages “because, just like anything that you’re starting for the first time, you need to crawl before you walk, and walk before you run.”
The crawl stage is where a company is still determining how they want to use analytics. They know that they collect data but are still working out where it can be used to predict growth opportunities.
At this stage, you need someone who can map out the path from collecting data to successfully using predictive analytics. This needs to be a strategist who can help you figure out how you want to use data and exactly where the data will be coming from. A top-notch mathematician or data scientist isn’t the correct hire at this stage. You need someone who can ensure that crucial data is collected and reported in the form of a data set. This person needs to be a mover who can instruct others on what to collect and how to present it. This person will teach you how to use your reported data to track all areas of your business whether, in marketing, sales, finance, and operations related. The goal of all of this is to take everything that happens each day at your company and use it to create readable data that can be analyzed and researched.
Now that you’ve mastered crawling, you’re ready to start walking, step-by-step. You need someone who can create reports about the new data your company is collecting. They should be highly skilled at creating easy-to-understand dashboards for you and the rest of the executive team.
If this stage is successful, you and your team of decision-makers should know the ins and outs of what’s going on in your business simply by reading a report that tracks your data. You should be able to look at these reports and be able to see what will help you capture the best return on all of your investments. You should also be able to determine from the reports the sales strategy that takes the smallest investment of time to produce the best results. You’ve got to the stage where instead of digging through all your data, hoping to find answers, you can pull up a dashboard that highlights your KPIs. When you feel comfortable enough with these reports and the insights they are providing, you have officially mastered walking.
The final stage is learning how to run. At this stage, you can use your newfound insights to make accurate predictions that lead to business growth. All the data you have been collecting and interpreting can now be used to build models or predictions.
It feels like a sprinkling of magic. Take some programming and advanced mathematics and stir well. Poof! Now you don’t only understand your customers, you can predict what they want or need in the future. Predictive analytics take past and current data, trains a data model, and comes up with highly reliable predictions about any aspect of your company.
The return on investment that you’ll see when you are finally up and running with predictive analytics is massive. Your product and service offerings, at any given time, will perfectly match the current needs of your customers. This is the point where your business has finally achieved “data-driven” status.
This is the stage where you need data scientists. Data scientists are analytically-minded, statistically and mathematically sophisticated data engineers. No wonder they are rare. The running stage is where you can become truly competitive in your field. The more models you build, the faster you’ll run.
So when you decide as a business leader to make the transition to “data-driven decision making,” make sure you’re asking yourself: Where are we right now? Are we learning how to crawl? Is it time to start walking? Or are we truly ready to run at this point? When you can accurately answer these questions, you can then position yourself to make the best data science hire for your company.