Creating an Operations Analytics Framework to Drive Business Improvement
by Ron Powell
Originally published March 21, 2018
Anu, your recent presentation focused on analytics and Teradata’s support for popular tools like Spark and TensorFlow. With these powerful artificial intelligence (AI) tools expanding within the enterprise, what is next for data science?
Anu Jain: I really believe we have reached a seminal moment in Teradata’s history and Think Big’s evolution as an organization – moving from a closed database to an open garden where we now can have the tools such as Spark, TensorFlow, R and Python used for new data science initiatives. It’s really about democratizing data science to the masses versus being a one-off science experiment.
How does an organization assess buy versus build strategies when it comes to analytics?
Anu Jain: As I have spoken to many clients over the last couple of years and especially in the recent past, the idea of analytics is becoming a vertical capability for many organizations. It’s becoming a strategic opportunity for them versus just an IT component. The first question is about the vertical capability for the organization. Are they going to compete on analytics and use that to drive a business process forward? Or is it more of a nice-to-have capability to help assist a business process? It really is about whether they need to own the IP and have that IP in house to drive profitability or if they can partner with someone who can have their data and be a supplier to them.
For analytics, we’ve been focused on data, engines, tools and technology, but what about business outcomes. Where are you seeing your clients successfully connecting their advanced analytic programs to derive business value for the enterprise?
Anu Jain: I think one of the most important considerations when we’re looking at analytics is that it’s not about the one-off data science experiment. I’ve seen a lot of what I call “Frankenstein” data science experiments that happen in a corner in a small lab, and they’re backwards looking. If a customer is trying to solve customer journey or predictive maintenance for an airline, it’s not a one-time model. It’s all about what we call analytic operations – taking that model and deploying it at scale across the business inside a workflow so that they’re continuously getting value from it. What I discuss with our clients is their analytics operations framework to determine how they can take the model, deploy it at scale, and then continue to monitor it to drive value versus a one-time experiment.
How do you track the success of an advanced analytics program?
Anu Jain: There are a number of things we look at. First of all, do they have quantifiable business outcomes? Does the capability drive something of true business value that can be measured? Secondly, it’s about productionalization of those results. Is it a one-time value or can it be driven into the workflow? Can it be scaled, and is it a continual part of their business process?
We are really starting to see that analytics is the business improvement lever that drives outcomes – not only measurable but also sustainable.
From an enterprise perspective, are these mainly older enterprises or newer companies?
Anu Jain: It is interesting. We are seeing many legacy clients who view analytics as pivotal to their corporate strategy. I was talking with a large aerospace manufacturer recently, and they see analytics changing their entire business model from being just a manufacturer of airlines to a services-oriented organization. We’re seeing it in the press today with companies like GE, where they’re betting the entire company on an analytics-based strategy.
Then we also have what I call the “new age” organizations – such as Netflix and Apple. They, too, see analytics as their Sherpa, taking them up the mountain to be successful.
Thank you, Anu, for your insight into analytics today.
Recent articles by Ron Powell
Copyright 2004 — 2019. Powell Media, LLC. All rights reserved.
BeyeNETWORK™ is a trademark of Powell Media, LLC