Top 5 traits of a good data scientist

Source:https://www.inc.com/jeremy-goldman/why-brands-agencies-brands-shouldnt-hire-data-scientists.html

“Try not to become a man of success but rather try to become a man of value”

The powerful statement once told by Einstein. Be it a data science world, or a manufacturing industry business, all run by people. In each of the world, several people got success and one thing is quite common in all of them. Trait to succeed runs into the blood of every successful person.

By 2020 in the US there will be an increase in data science professional jobs from 364,000 to 2,720,000. Not only that, the annual demand for such professional jobs will reach nearly 700,000 by this year.

Companies virtually in each and every industry looking to extract the information about past, present, and future from their existing data, and thus, data scientist jobs will continue to be in high demand.

But not every data scientist is a great or very successful data scientist. What differs a great data scientist from others.

1. A good Storyteller

A good data scientist's job is not only to develop a prediction model or analysing the data to its core but also to tell a compelling story to non-technical management, client, or team members about the analytical solution which he/she developed. For example:

Source: https://www.kaggle.com/ash316/eda-to-prediction-dietanic

If a data scientist has presented large chunks of numbers and try to explain how age and gender or passenger class are related to the Titanic survival, he would be failed. But the above graph clearly demonstrates all the relationships, which would be easier to explain through a story.

2. Thirst to learn

Try to learn something about everything and everything about something.

It is appropriately told by Thomas H. Huxley and the context would be aptly suited for a good data scientist.

Source: https://www.simplilearn.com/data-science-vs-data-analytics-vs-machine-learning-article

Technology is advancing on the top gear. One would be leftover if it stops learning especially in the field of data science. XGBOOST, LGBM, CATBOOST like advanced machine learning algorithms are not known before 2017 and now they are one of the biggest contributors to predictive modeling.

In addition, one can infer from the deep learning timeline that if a good data scientist has no thirst to learn, he would be obsolete in this data science world.

Source: https://www.kdnuggets.com/2018/03/weird-introduction-deep-learning.html

3. Business Mindset

“If I were given one hour to save the planet, I would spend 59 minutes defining the problem and one minute resolving it,” Albert Einstein said.

Source:https://businessuntangled.com/2020/03/02/do-you-have-the-mindset-for-home-based-business-success/

“Have you understood our problem? Can you find the solution?” Every client meeting with the vendor ends with this statement.

In the data science world, if the data scientist will not have a Business Mindset, he would fail to ask a lot of questions related to business and problems. It will result in a solution with multi-fold assumptions. Thus, it will result in an unhappy client and a failed analytical solution. A good data scientist goes deep dive not only into the problem but also into the business and their direct and indirect impacts on the analytical solution.

4. Solution Finder, not a mechanic(knows a lot of tools)

Source:https://execleadercoach.com/2018/02/21/solution-finder/

It comes as no particular surprise to discover that a scientist formulates problems in a way which requires for their solution just those techniques in which he himself is especially skilled.

While explaining the law of the instrument, Abraham Kaplan has stated the above quote. While Moslow has coined it better:

“If all you have is a hammer, everything looks like a nail

A data scientist should thrive for:

  1. understanding the complete business problem
  2. understanding the risk and current status of the business problem
  3. looking for the requirements to solve the problem
  4. lastly decides about tools and techniques

Client/Customer/Stakeholder cares least about the tools and techniques. They mean business and they need an effective solution. They hardly care if you solved the problem using R or Python. They hardly care if you solved the problem using logistic regression or deep learning. A good data scientist also looks for the solution first, not focus only on tools and techniques.

5. Out of box thinking — creative

Source:http://www.keypersonofinfluence.com/how-can-we-think-outside-the-box-from-inside-the-box/

“My model is not performing well until I have created new features from the existing features. I got amazing results now”

“Regular implementation of the model is not giving the desired result. I have created an innovative stacked model.”

“I was thinking of how to present the result to the client, as it is too much data-centric. Lastly, I pulled out an amazing visualization to tell my result to the client”

“This is an amazing finding, but how to show to the client in an effective manner. Oh, I got an interactive chart”

To detect breast cancer, if the features could tell a story using swarm plot
To detect breast cancer, if the features could tell a story using swarm plot
Source:https://www.kaggle.com/kanncaa1/feature-selection-and-data-visualization. The swarm chart is showing how different features could play a role in detecting breast cancer

If you are a good data scientist, you have to think out of the box, creative solutions not only to solve problems but also to present them effectively.

Raw data is like blank paper with colors, you have to be a skilled painter to portray it into an art.

Conclusion

To earn the badge of one of the greatest data scientists, one has to do more than routine work. One has to be passionate about data science and not only learn different tools, techniques but also have to be creative. One has to be a very good data storyteller. It is not required to create a complex model to solve the problem, but to understand the data and the problem from a business perspective and giving try to create features from the data can help in solving several complex problems. A good data scientist always starts with a simpler model.

Senior Data Scientist in TCS, more than 10 years experience in Data Science, Top 20 Data Science Academician in 2018.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store