Meet Sushma Vasudevan, VP of Analytics and Data Science at PureSpectrum. Today Sushma shares her passion for data science and how it led her to be a co-creator of PureScore™, PureSpectrum’s award-winning respondent-level scoring system.
Q: How did you first get interested in data science?
A: Growing up in a family of mathematicians, our dinner table conversations often turned into something quantitative. As a child, I was curious to understand everything from a mathematical perspective.
In undergrad, I was formally introduced to statistics and it was instant love. The application sciences and theoretical math just came together for me. My first job wasn’t in data science but as a quant analyst for marketing.
Then the digital era began and data was everywhere. It felt like there was a data explosion happening. As data became Big Data, suddenly there was this new thing called data science where statisticians and coders came together. And I wanted to be a part of it. So I went back to school to get all the skills needed to become a data scientist.
My first job as a data scientist was at IBM where I got to do some very innovative projects like the net promoter scores and marginal analysis for revenue generation. It may sound dry to some, but I loved it and knew I had made the right decision to go into this specialization.
Q: How did that lead you to PureSpectrum?
A: Besides working at IBM, I was also teaching statistics and market research at California Lutheran University. I invited Michael McCrary to speak to one of my classes. Michael is the entrepreneur in residence for CLU and he’s a bit of a celebrity on campus. People know him because he started PureSpectrum at Hub101, the university’s entrepreneur incubation organization.
Michael gave an incredible lecture to my class and afterward, we got talking. He told me how he planned to revolutionize the market research industry. He too saw the huge data explosion happening but he recognized the gap in the market. He saw that PureSpectrum had the potential to change market research technology and make life easier for researchers. His goal was to create a data-driven and innovative company, but keep it simple and easy to use. Eventually, Michael introduced me to Mark Menig and those first conversations with Mark sealed the deal. I knew PureSpectrum was where I was meant to be.
Q: You are the mind behind PureScore™, what’s the origin of this technology?
A: When I joined PureSpectrum, my first responsibility was to redesign PureScore™. At the time it was a rule-based data scoring system but Mark had a grand vision for it. So with his breadth of MR industry expertise and my depth of data science knowledge, PureScore™ saw a new light. As it grew and developed, it went from a rule-based scoring to the data-driven, model-based, machine learning algorithm it is today.
And to be honest, that’s why it’s so successful. PureScore™ is the culmination of Mark’s understanding of the industry and vision for data quality combined with my technical and scientific curiosity. I’m so proud of the algorithm we’ve “raised” together. It’s one of the most innovative things that I’ve ever built. But we aren’t done. We are always thinking about new ways to improve PureScore™.
It’s been two and a half years and it’s never stopped growing. I believe that we are creating a benchmark for the industry. I want PureScore™ to be a minimum threshold that every respondent has to go through. As sample buyers’ expectation for quality respondents increases, PureScore™ will make sure that they always reach them.
Q: What else do you think sets PureSpectrum apart from competitors?
A: In my perspective, innovation is when you take complex concepts and technologies and present them in a manner that anyone can understand. It’s easy to throw big words and industry jargon, but ultimately, what is the value-add for your end-user if they don’t understand the tools you are creating for them?
Innovation sets us, PureSpectrum, apart from the competition. We call all our data-science-driven assets, PureSolutions. Of course, we have PureScore™ which is our respondent scoring system. We are also in the process of rolling out PureText™, which is our natural language processing technology. And after that will be PurePrice™, our pricing recommendation algorithm. All of our solutions take millions of data points and lines of code. But the PureSpectrum difference is that ultimately when you look at it from the front end, everything is so simple. And that’s what we are always striving for, to make life easier… for clients and respondents.
Q: What is a common misconception people have about data quality?
A: That data quality is a series of checks for straight-lining and red herrings. When I first started in this industry, every supplier had very similar quality check; standard questions, system-level de-dupes, etc. But I did not hear anything about inherent data quality. What if a respondent maintains two different personas with two different suppliers on two different platforms and responds to the same survey? Isn’t the quality of the data collected diluted inherently? And that is exactly what PureSpectrum addresses as well. Our platform has visibility into multiple supply channels and we have the technology to identify a respondent coming from different suppliers. So data quality, for us, is not just a check box, it is our priority.
Not to brag but I think PureSpectrum has the best tools to approach data quality, no matter how dynamic the requirements. Especially in the MR industry, quality is a cat and mouse game. We put in a check and someone finds a way around. We put another check and the fraudsters innovate and morph. So, we must stay ahead in this game at all times. And that’s what we do – continuously innovate and lead the charge against fraud.
Q: What other passions or hobbies do you have?
A: My mind is constantly moving and as a result, I tend to get bored of things quickly. So, I have a lot of hobbies and interests but there are three things I really love to do. Number one is music. I’m a trained musician. I love to listen to and compose music. I have a 19-year-old daughter who is also a musician and a gifted painter. We love to sit and discuss art, for hours, together. It is my go-to, especially when I’m stressed out.
My next interest and my most recent passion is cryptocurrency. I love learning about crypto and the blockchain technology that works behind it. It intrigues me so much. I follow the crypto market ardently.
And my third passion will never go out of fashion! Binge-watching TV. Mostly, crime dramas and thrillers. That is another stress buster for me. I feel so happy when I do all these things. They all challenge my mind in different and creative ways.
Interested in learning more about PureScore™ and PureSpectrum’s other innovations that can make your life easier? Connect with our team below.