It pays to be a data scientist-literally and figuratively. According to the founder of LinkedIn, data science-based jobs grew 15 to 20 times in a three-year period. And earlier this year, LinkedIn named data science “the most promising job of 2019”.
With a median base salary of $130,000 and 56% more open roles this year than last, it’s not hard to see why this job’s attractive. But for many businesses, it’s also out of reach.
Most people can’t build the algorithms required to power predictive analytics, data mining, and other big data techniques that businesses need. That’s why data science demands high levels of advanced education in niche areas of math and computer science.
This leaves huge demand for people with such skillsets - and not nearly enough applicants to meet that demand.
Luckily, not all is lost for businesses who want to use data science but can’t spare the salary. Demand for data scientists is growing in tandem with the amount of business intelligence (BI) tools. Augmenting data science by combing the right software and employees creates a new, equally crucial role: Citizen data scientist.
Citizen data scientists are software power users who can do moderate data analysis tasks. They don’t replace expert data scientists: Instead, they use software features like drag-and-drop tools, prebuilt models, and data pipelines to create models without code.
Unlike expert data scientists, you don’t need to recruit citizen data scientists. That’s because they’re already part of your organization. Citizen data scientists know the ins and outs of your unique business: Its strategy, customers, pain points, and tech stack.
This knowledge helps citizen data scientists bring business and industry acumen to their use of advanced data and analytics features . These skills complement the deep domain clout in math and computer science that expert data scientists have. Indeed, people in both roles will see the most success when they work together.
Expert data scientists collect and clean a wide range of data, then analyze it to solve business problems. The data they analyze ranges from structured (which basic algorithms can easily search) to unstructured (which includes more human language and is tougher for algorithms to analyze).
Many data scientists also build the algorithms that they use to analyze structured data. They also use external tools (like Elasticsearch and Cluvio) to analyze large sets of unstructured data, like text within emails or freeform survey answers.
This requires data scientists to make search queries for specific words and phrases, then manually analyze unstructured data. As such, data science has high barriers to entry.
Most companies require experience with statistical programming languages, like R and Python. They’ll also want knowledge of third party data analysis tools like Google Analytics, along with a Master’s or PhD in Math, Computer Science, or another quantitative field. Most expert data scientists (88%) have at least a Master’s degree, and nearly half have PhDs.
Citizen and expert data scientists bring some skills that overlap, and others that are unique in their own rights. Ideally, you’ll have a big enough budget to find citizen data scientists in your business, then hire expert data scientists to take citizen data scientists’ work to the next level.
But before you put the job description up, confirm what your business needs are - both in the short term and five years from now. Gartner’s latest Hype Cycle for Data Science and Machine Learning shows several stalwart techniques in the Trough of Disillusionment.
From predictive analytics to deep learning, many of the sector’s trendiest techniques creep dangerously close to disappointment. Without a clear reason for fitting these techniques into your business strategy - and enough clean data that’s ready for analysis - hiring an expert data scientist won’t deliver the results you need.
Not sure where to start? GetApp’s infographic shares the key differences between citizen vs. expert data scientists.