Data Science

This is how a typical day of a data scientist looks like

17th Mar `17, 11:03 AM in Data Science

Data drives companies’ success. It is very important to manage data because it runs systems, businesses, academies and…

Audrey Throne
Audrey Throne Contributor
Follow

Data drives companies’ success. It is very important to manage data because it runs systems, businesses, academies and dialogue. We can predict natural disasters, improve social media reach, dethrone anarchists and sell more stuff using data. But how exactly do we use bits of information to make them useful for us? The answer lies in big data.

Businesses today need to do more than merely acknowledge big data. They need to embrace data and analytics and make them an integral part of their company. Ronald Van Loon

So what exactly is big data? It is a term used to describe sets of data which are so large that a single software may not able to analyze it all. The amount of data in today’s world is literally exploding, and to analyze and utilize these sets, there has to be more than one system involved.

That’s where a data scientist comes in. These professionals are skilled to manage data for businesses and institutions and turn big data into important insights. This job is often mistaken as that of a Data Analyst’s, but these are two different things.

Data Scientists are data-driven individuals who do it more for their passion than for the paycheck. The confusion between data analysts and data scientists wrongly leads companies to fill this gap by taking developers and analysts onboard, thus missing out on the benefits that come with working with a data scientist. Britain is expected to create an average of 56,000 big data jobs a year until 2020. With big data talent in short supply, companies are increasingly willing to pay sky-high salaries to bring in the right skillsets, with many individuals commanding six figures. McKinsey & Company reports that by 2018, there will be 140,000–190,000 data scientist job postings that go unfulfilled. Worse (for the employers), there will be 1.5 million managers needed to optimize available data. Therefore, the next three years offer a veritable goldmine for data scientists.- The Guardian

So, what do data scientists do? On an average, they work 60 hours per week. To tackle the daily flow of big data for their institutions, these professionals  master certain skills in their daily routines. We have added some quotes from professionals who do the drill everyday:

Fix problems:

Since I work on a team that focuses on reporting and data quality, if there is a new product out there we might want to incorporate that into one of our major dashboards so that will involve working first with product mangers to work out what is important to the product, engineers to make sure the relevant data is being tracked, and then working with our data services team to do ETLs (extract, transform, and load) and visualizations. – Abraham Cabangbang, Senior Data Scientist at LinkedIn

The major part of a data scientist’s day is spent in fixing problems, rewriting code and ordering subordinates to do something correctly again. At the same time, they create the groundwork for future projects and then go home to work on the projects until they fall asleep.

Attend meetings:

Data scientists need to attend several meetings each day, most of which are to run checks on ongoing projects and campaigns. Dr Lionel Raymond, Data Scientist at a Fortune500 company expressed on Reddit how his typical day involves so many meetings:

“Went to meeting. Told C level management that a project they were hoping I’d be able to tackle would be pushed back an undisclosed amount of time. Why? Because the 18 months of historical data I was promised doesn’t exist. “Well how can that be?” they ask. I don’t know. Go ask the DBA that set the data to overwrite the existing tables every month rather than append.

Went to another meeting …to ask them how to expect the data science team to increase revenue by $10 million next year when getting approval for additional RAM on a mission critical server has been sitting in processing for the last three weeks. The $200k to cover promotional knickknacks related to a completely unnecessary rebranding of a platform we are replacing in less than a year and have already migrated 98% of our clients off of flew though the department in just a few hours though.”

One major problem these professionals face is to make other people understand what they do and why they ask for a certain piece of information.

Handle more than one project:

Every data scientist agrees that they can comfortably handle more than one project at a time. A lot of automation is needed to manage several projects simultaneously.

Lots of phone calls are made every day, and the best thing is that you can run up to the person who sent you a design of something and ask them what the data meant. It requires frequent communication. Many experts prefer time-tracking software like Basecamp, to help them stay on top of multiple tasks and free of clutter.

I like to work on more than one project, so I would be involved in a number of meetings or design discussions. A lot of my day is spent in phone calls talking with others. [You can talk to the person] who designed something, and they know what this column of data means. Then I keep up with videos and course work and go to as many seminars as I can. I live in the South Bay, and all the good seminars tend to be in this area.- Ram Narasimhan, Data Scientist at General Electric

Analyze data:

Analytical skills are a must for a data scientist. A large chunk of their day is spent on analyzing various sets of data and categorizing them. Sometimes they take raw data and turn it into something useful with their professional analysis. Then they put this analysis into a context to find a new strategy for the business.

Handle their own business domain:

Data scientists get really bothered when they don’t know all about something. The topmost thing a data scientist perfects is the knowledge of the business they work with. Since every business requires a data scientist, the workflow for each professional differs. Not only do they require constant updates from their employers, but they also want to stay on top of the key metrics that matter to the business.

Presentations:

As with any high activity job, a data scientist has to juggle with a lot of presentations during the day.  The results acquired from categorizing and analyzing data need to be presented to the stakeholders in a very efficient manner.

A successful data scientist presents the results of a data analysis, strategies for upcoming, sales methods and other measures to the stakeholder and listens to their feedback at the same time. This way, they learn more about the business problem and concentrate better.

Wrangle the data:

Companies hire a data scientist when they have a lot of data and nobody to make sense out of it for them.  This is particularly easy, since tech tools and analytical skills help in sorting out the chaff easily. In this type of job, the professional makes meaningful data-like contributions to the code and provides basic insights.

Do the math:

Data scientists have to be excellent at mathematics, statistics and all things analytics, combined with top-notch technical and communication skills. Knowledge of the latest trends help them handle the data using the right software every day.

MORE FROM BIG DATA MADE SIMPLE