Data Analytics vs. Data Science: Deciphering the Differences
You’ve probably heard the term “big data” by now. With more tools to collect data than ever before, companies are racing to utilize the goldmine of information they have access to. In fact, some experts believe over 90 percent of the world’s data has been created since 2011.
User information, consumer information and performance metrics—the categories of raw data are seemingly endless, but getting useful information out of all that data is easier said than done.
This is where data analytics and data science come in. Careers in these fields are skyrocketing, but even people well-versed in technology can get confused about the job titles. So what’s the difference between a data analyst and a data scientist? We asked current experts in data to help clear up the confusion.
Data analytics vs. data science: What’s the difference?
These two disciplines both involve organizing massive amounts of data to come up with measurable, usable intelligence. They both have a problem-solving and critical-thinking core, while also involving a highly technical skill. So what is the difference?
“The most important difference, in my opinion, is that data analysts answer questions that have been asked using a framework that has been established,” says Maya Wilson, data scientist at Axio. “Data scientists come up with the questions and the methodology to answer it.” As an example, a data analyst might be tasked with finding answers to a specific question or set of questions. A data scientist might try to predict future questions and find potential future uses for the data.
Data analysts typically work in close communication with the company in a direct strategy to make improvement. “Working as a data analyst usually involves engaging with a wide variety of people, who are all employing data to guide their decision making and inform their business strategy,” says Johannes Mehlem, data analyst at Hubspot. “There is a lot of satisfaction for me personally in being a ‘truth-seeker’ about what is really going on in the business and being a reliable resource to the company.”
“Data science is more open-ended and requires creativity and theorizing to know what kind of questions can be answered using data, what kind of data will be needed to provide meaningful evidence, and what kind of insights are relevant,” Wilson says.
Data analytics vs data scientist: Daily tasks
The differences between these fields can be pronounced. But what does that mean on a day-to-day basis? Data scientists and data analysts are responsible for specific tasks in their days.
A data analyst’s day
“First, I regularly check our ongoing reporting in regard to key metrics such as conversion rate (the percentage of people who take a certain desired action on the website),” Mehlem says. “In case of any abnormalities, I will dive deeper into the data to understand what has caused a significant change. This is usually done by segmenting the data (e.g., Were there changes to the traffic volume, traffic composition or any on-site changes such as change of copy or design that explain those significant changes in conversion rate?).”
Mehlem’s next task is to spend time thinking about how to improve data quality and comprehensiveness. Asking questions like, “Are there other ways to measure how engaged a user on the website is?” can help you to improve the data quality on a website. After this, Mehlem meets with decision makers to provide analysis so they have more data to act on. “Also, I meet with marketers to discuss their business goals and to understand how the analytics function can ideally support them in measuring success.”
A data scientist’s day
“For example, during my meeting today, we discussed what kinds of metrics we can gather from our user’s behavior or what kinds of demographic questions we would need to ask to be able to make meaningful comparisons between users,” Wilson says. “I’m on the software development team so I spend time doing more standard development stuff as well, such as testing my colleagues’ work, looking for bugs in our product and training new developers.”
Data analytics vs. data scientist: Important skills
We took a closer look at what skills employers prioritized in their job postings for data scientists and data analysts in the last year. Here's what we found:
|Data analyst skills*||Data scientist skills*|
|Microsoft Excel||Machine learning|
|Microsoft office||Apache Hadoop|
Data analytics vs. data scientist: Education requirements
The industry is still getting used to these two separate terms. But employers are out there now searching for both data scientists and data analysts. We used real-time job analysis software to scan through the online job postings for data scientists and data analysts in the last year to see what they are looking for.
What we discovered was that the vast majority of data analyst jobs (nearly 85%) require a Bachelor’s degree.1 The average required education level is a little higher for data scientists. Fifty-six percent of job postings for data scientists asked for a Bachelor's degree, with the remaining jobs looking for a Master’s degree or PhD.1
“Data scientists need more training than data analysts because the job requires more critical thinking,” Wilson says. “Undergrad teaches you how to ingest and analyze facts and information. Graduate school teaches you how to think critically and find weaknesses in arguments and ideas.”
Wilson goes on to say that data scientists need to invent and criticize methodology to stretch what data can do—as well as to figure out what data can’t do. “In summary, applying advanced and sophisticated data modeling techniques is highly technical and requires training in statistics, data management and critical thinking.”
Data analytics vs. data scientist: Average salary
Given the relative “freshness” of these two careers, there isn’t a lot of specific salary information out there. That said, the Bureau of Labor Statistics reports a 2016 median annual salary of $81,950 for mathematicians and statisticians—a career category with plenty of overlap.2 For a more granular salary comparison, we used real-time job analysis software to compared the advertised salaries for data analyst and data scientist job postings.
Of the 2,024 data scientist job postings that shared salary information, the median salary was $135,000.2 On the other hand, of the 6,352 data analyst job postings that shared salary information, the median salary was $62,000.2
As any good data analyst or scientist would note, these figures are not necessarily a complete picture of average wages—not many job postings list salary information. That said, the overall trend of data scientists earning more than data analysts holds true as the skill, education and experience requirements for data scientists are substantial.
Data scientist or Data analyst: How do you choose?
As you can see, these jobs can be very different. But they both still involve making sense of massive amounts of data to gain insight and make improvements. If you’re torn between the options, consider your own personality.
The analyst side
“My decision to work in data analytics was really based on role-matching behavior that I exhibited already before entering the professional world,” Mehlem says. Mehlem knew he was good with numbers, had an analytical mind and actually enjoyed deeply focusing on one issue for many hours.
“I would start with self-reflection on whether or not you have inclinations toward numbers, prolonged phases of focus, communicating complex issues in a simple manner and curiosity.” Mehlem says traits like these can definitely be learned, but that it’s a good sign if you have them already.
Mehlem also points out that while pulling insights from data is very rewarding, it takes a lot of sustained work to get there. “In reality, most data analysts will spend significantly more time cleaning data than presenting the insights from it. Ask yourself whether you’d lose momentum when sifting through data for weeks or even months or are likely to enjoy the process of working toward something.”
The scientist side
“I think it’s important to assess how comfortable you are with uncertainty,” Wilson says. She advises students ask themselves these questions: “Are you prepared to start from scratch doing something that has never been done before? Can you make a case and defend your choices against alternatives? Can you acknowledge and own the limitations of your strategies and work to continuously improve your methods?”
“If so, data science will be a fun challenge,” Wilson says “But it’s also hard to work with so many moving pieces and can leave you feeling like you never know enough to do your job as well as you’d like.”
On the other hand, if you prefer being asked a question and hunting down the right answer, you might be better-suited for a data analyst position. Wilson says becoming a data analyst offers more structure and might feel more rewarding for people who prefer working with black-and-white questions and answers.
The wide world of data
Though our current abilities to gather and clean data are pretty recent developments, the industry of data is absolutely booming with growth and potential change. Data analysts and data scientists aren’t the only careers that specialize in data, and more job titles are likely to spring up as time goes on.
But if you think you want to work in data, you don’t have to make any final decisions right away. As you say above, the majority of data analyst and data scientist job postings ask for aBachelor’s degree. You can begin your new career path in data without deciding between the two immediately.
If you would prefer to get more information about what one of these jobs is like from the inside—check out our article, “What Does a Data Analyst Do? Exploring the Day-to-Day of This Tech Career.”
*Burning-Glass.com (analysis of 35,327 Data Analyst and 12,388 Data Scientist job postings, Oct. 01, 2016 – Sep. 30, 2017)
**Salary data represents national, averaged earnings for the occupations listed and includes workers at all levels of education and experience. This data does not represent starting salaries and employment conditions in your area may vary.