Data Science and Machine Learning: From Academics to Economics

Reading Time: 3 minutes

In my previous post, I shared our excitement that TIBCO’s analytics solutions received the highest score for Production Refinement in Gartner’s Critical Capabilities for Data Science and Machine Learning Platforms 2018 report. To recap, the Production Refinement use case is heavily influenced by the Delivery, Model Management, Performance and Scalability, and Machine Learning critical capabilities. Furthermore, Gartner stated that production refinement is where Data Scientists spend most of their time.

Most organizations are keenly aware that analytics is the fuel for a successful digital transformation. There are all sorts of studies and statistics that suggest that digital leaders significantly outperform digital laggards by a significant margin (e.g. 5x higher revenue growth) with only a slight increase in technology spending (3.5% vs. 3.2% of revenue). So, given the fact that many organizations have the access to voluminous amounts of data and there are a plethora of algorithms available, is there a secret ingredient that most organizations are missing?

Let’s use Netflix as an example. On a daily basis, Netflix streams 250 million hours of video around the world (190 countries) to 98 million paying subscribers. Everytime you click, browse, watch, pause, rewind, stop, start, rewatch, all of this data is collected and analyzed to personalize your next experience. We are all keenly aware that companies like Netflix, Amazon, and Google use recommendation systems and make next best offer recommendations; but did you know that Netflix even personalizes the graphics for each viewer segment? For the show “Stranger Things”, did you know that I may find the show in the “TV mysteries” category and you may find it in the “sci-fi thriller” category? Netflix has around 2000 “taste communities” (aka customer segments) and for the popular series “Stranger Things”, Netflix’s algorithms apply 12 tags to capture the intricacies of how different people relate and react to the show; in other words, there are 2000 different customer segments each with its own unique viewing experience and 12 different intricacies for this particular show! Now, that must be a lot of AI, machine learning, and analytics being used to optimize your viewing experience.

Netflix’s runs on Amazon Web Services (AWS) and consists of a whole lot of infrastructure, microservices, and data science to stream shows around the world. Netflix’s ability to orchestrate people, processes, and analytic technologies in real time has allowed them to monetize their data and gain a significant competitive advantage. This is the secret ingredient that makes Netflix who they are. Now you don’t have to be Netflix to pull this off.

We have many customers around the world who are managing hundreds and perhaps thousands of models. Whether they be in banking, insurance, manufacturing, energy, or health care we hear time and time again, the ability to put quickly test and deploy analytics to production systems is what leads to competitive advantage. Through automation and orchestration, one customer in the banking industry reduced the time they spent on developing and deploying models by 50%! That 50% more time they can be innovating and optimizing customer experiences.

When organizations think about data science, their mind quickly races to the algorithms and mathematics. However, this is not where the real value lies. The real value lies in being able to monetize their data through analytics which create insight. To deliver the insight, companies need to increase focus on the deployment, management, and monitoring analytic models. In order to move from academics to economics, organizations need to focus on production refinement, this is the secret ingredient which will lead to competitive advantage.

Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner’s research organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.

1 Harvard Business Review. What the Companies on the Right Side of the Digital Business Divide Have in Common. By Robert Bock, Marco Iansiti, and Karim R. Lakhani. January 31, 2017.

2CNet. ‘Stranger Things Addict? Here’s how Netflix sucked you In. October 23, 2017.

3How Netflix works: the (hugely simplified) complex stuff that happens every time you hit Play.

Previous articleF1 Race Recap: Heartbreak at the Austrian Grand Prix
Next articleWhat’s a “Data Function” and Why it is Such an AI Enabler?
David Sweenor has close to 20 years of hands-on business analytics experience spanning product marketing, business strategy, product development, IoT, BI, and advanced analytics, data warehousing, and manufacturing. In his current role as the Global Analytics Marketing Leader at TIBCO, David is responsible for GTM strategy for the advanced analytics portfolio. Prior to joining TIBCO, David has served in a variety of roles—including a Business Analytics Center of Competency solutions consultant, competitive intelligence, semiconductor yield characterization, data warehousing and advanced analytics for SAS, IBM, Quest, and Dell. David holds a B.S. in Applied Physics from Rensselaer Polytechnic Institute in Troy, NY and an MBA from the University of Vermont.