Organizations everywhere want to crunch their internal data for crucial insights that will determine their strategies for years to come. As a result, data scientists with the right mix of skills and experience have their pick of employers. How much can you earn as a data scientist?

According to levels.fyi, which crowdsources compensation information from tech professionals, some companies are more than happy to pay data scientists over $500,000 per year once you factor in all compensation streams, including equity and salary. Here’s a breakdown of the top-paying companies:    

Not everyone can work for Netflix or a major financial firm, of course, but data science can still pay well even at smaller organizations: Business Insider recently analyzed data from the U.S. Bureau of Labor Statistics and pegged the median 2022 compensation for data scientists at $103,500 per year. What’s more, the profession is expected to rapidly gain jobs over the next several years, alongside other tech professions such as software developers, information security analysts, and others.

Of course, unlocking that kind of pay hinges on mastering key data science skills such as Python, SQL, Tableau (and other visualization tools), machine learning, and much more. That’s in addition to mastering data science disciplines such as:

• Gathering the data: Data scientists usually start by gathering data from different data sources. This step is often known as extraction.
   
• Cleansing the data: Although we’d like to believe that the data we receive doesn’t have any errors, in reality many things can go wrong that will cause data to be lost or incorrect. And that means the data must be cleansed. Bad data is either removed or corrected; missing data is either ignored or corrected.
   
• Transforming the data: The data might not be in the form you need it. For example, weather data might come in as Celsius but you need to convert every item to Fahrenheit. Or the data might come in grouped as sales orders for each customer, but you need just sales dates and quantities for each individual product. This requires going through the entire dataset and reworking it into a form you need it in with the help of automated tools meant for processing large sets of data.
   
• Joining the data: This is technically still part of transformation, but in this step, the data scientist connects data points from multiple sets (such as car diagnostic points with the current weather when the problem occurred).
   
• Loading data: In this step, after the data is in its required form, it gets saved in some place, typically in the cloud if it’s large. (And notice now this data can also now be used in future projects as a new source of data.)

On a recent episode of ‘Tech Connects,’ Shadi Rostami, SVP of engineering at Amplitude, also discussed the rise of “data democratization,” which is the ability for employees throughout the organization to gather and analyze data without much training or assistance from data scientists, data analysts, and other experts.

Even if you’re not a data scientist or analyst, learning data science tools (which are becoming easier to use) and becoming more data-literate can translate into new and exciting career opportunities, particularly if your current or future employers are trying to analyze massive datasets.