Podcast

Full Episode Transcript:

Hi and welcome to hello, Human, a podcast to explore ideas and feature humans working in AI and technology.

Jon: Abhinav Kolhe and Sudhakar Pemmaraju from Cognizant join us today on the hello, Human Podcast where we discuss the latest topics in artificial intelligence and how it’s being applied in the real world. I’m Jon Knisley, the host of hello, Human and a long-time technologist.

A big thanks to FortressIQ for sponsoring the program, and be sure to subscribe wherever you listen to podcasts. On this episode, we’re going to explore the building blocks to developing a resilient digital core, and how automation helps businesses manage through crises. Welcome, Abhinav, and Sudhakar to the program.

First off, Cognizant is one of the world’s leading professional services companies. With more than 200,000 employees globally, you benefit from seeing how a lot of different companies in different regions and different industries have responded to the challenges created by the pandemic.

Maybe you can start by giving us a background on your roles and how the automation and transformation market has shifted due to the current environment. I hate to call it the new normal, I’ve done my best to avoid calling it that, but all indicators suggest that this is the way it’s going to be for a while. Sudhakar, do you want to start?

Sudhakar: Hello, Jon. Glad to be on this podcast, Jon. I lead the North American Digital Operations and IPA Advisory Practice at Cognizant. I’ve been with Cognizant for the last four years, and I’ve been in the advisory consulting practice of strategy and operations all my life.

Jon: Great, Sudhakar. Abhinav, you’re a bit more focused on automation in your work, so I’m interested to hear how your perspective on the impact of the current crisis and how it is shifting, thinking on automation among your clients.

Abhinav: Hey, Jon. Thanks for having me. I have been part of Cognizant for 20 plus years now and lead the technology office for Intelligent Automation practice at Cognizant. As part of my current role, I am responsible for practice development on advanced technologies like AIML, BPM, analytics, mining, and discovery tools.

To answer your question straight, I lately first acknowledged that new automation technology can address all the business impacts of COVID-19. We have seen in the recent past the way things have changed. It feels as if we have all done five years forward in this pandemic from a digital adoption standpoint.

The transition looks so very visible to all of us. If you look at banks, we see how smoothly they have transitioned to the remote sales and services teams, and how they have launched digital channels to engage their customers for making flexible payments of debt loans and mortgages.

Another popular example is our schools. Our kids go there. Schools have devoted 200% to online learning and digital classrooms. Doctors are another example in our common life. They began delivering telemedicine aided by more flexible regulations. These are just a few examples of how companies and businesses are working in the current crisis.

Two or three important things I would like to highlight as to how they would further pivot their journey. First of all, companies will need to ensure that their digital channels are on par to succeed in the current environment. Secondly, I think as the economy goes back in the next few months (I would say probably) demand recovery will be unpredictable. There would be the uneven spread of recovery across geographies, across products, customer segments, and whatnot.

Essentially, this will complicate matters for leaders as they grapple the way to deal with an uneven recovery that they are going to see. Historical data and forecasting models that they may have built in the past may not actually work because they don’t know how uncertain the recovery could be.

New data models have to be built. New analytical engines have to be built which will make life easier for operations. Essentially, to answer your question, automation-first or digital-first mindset will be an absolute essential center stage that companies will have to prepare for. We are seeing an upsurge from a lot of customers in terms of Hyper Automation use cases. Questions around what new platforms and toolkits should we actually bring to the table to deliver Hyper Automation use cases.

Jon: With all the buzz around Hyper Automation, I believe it was Gartner that labeled it the number one strategic technology trend for 2020. As a global service provider, how do you characterize Hyper Automation, and more importantly, what are the different components that a leading enterprise should be utilizing?

Abhinav: Yes, absolutely. In a very simple term, I would say Hyper Automation is end-to-end automation, accomplished by harnessing the power of multiple technologies from RPA, to machine learning, to AI, to process mining, and discovery tools, which essentially enables automation for virtually any repetitive task, as well as cognitive tasks for business users.

Hyper Automation goes beyond deploying bots for individual tasks. It is no longer just task-based automation. We are talking about a connected enterprise-wide change program that connects multiple teams, multiple work streams across an enterprise. The Hyper Automation platform consists of many tools. Some of the tools could be BPM platforms like BPM Low- and No-code platforms, like power automate as an example, or Unqork, Mendix type platforms. Process mining discovery tools like FortressIQ solenoids, AIML tools, and platforms. All of these toolkits integrated and stitched together in a common fabric. That’s what the Hyper Automation platform would essentially look like. 

One of the most key attributes of that platform would be the ability to loop in humans in the process. We call it human in the loop but typically, that’s one of the most important attributes of the Hyper Automation platform as well.

Jon: Tying this concept of Hyper Automation to resiliency, Sudhakar. I saw a report in Harvard Business Review a few months back that noted companies that had a better understanding of their processes, that had taken time to map their workflows, were less impacted by supply chain disruptions during the early stages of the global shutdown. They knew exactly which suppliers, sites, parts, and products were at risk which allowed them to put themselves first in line to secure constrained inventory and capacity on alternate sites. 

I’m curious if you’ve seen other business areas where Hyper Automation has supported resiliency.

Abhinav: Absolutely. I can give a couple of examples. Recently, a leading US alien company used virtual agents, human agents to quickly implement a way to process all refund requests caused by flight cancellations at the onset of the pandemic. Speeding the resolution, saving time, lowering cost, and more importantly, improving customer experience.

An insurer applied the NLP techniques along with the smart data ingestion tool, using scanning technologies to analyze policy terms to assess the impact of COVID. This allowed them to capture the data, extract it, interpret the information from documents, deal with cancellation requests, change the coverages, and address increasing transaction volumes.

Obviously, this has made the customers happy and improved customer experience. One more example is a large bank in India used NLP and AAP solutions to implement a government-mandated loan amortization program. This gave the bank’s customers the option to delay the loan and credit payments.

These solutions were actually used to rewrite loan agreement documents and communicate with customers. Doing so, the bank was actually able to reduce call volume significantly by diverting the most frequent request to cell service channels, to online FAQs and forms.

Now coming back to the supply chain. Predicting supply chain disruptions, so we have seen now this two-step process has a measurable operational impact during the pandemic. One leading US-based pharmaceutical company, we worked with used process mining, and machine learning to predict supply disruptions, automating alerts for inventory management and monitoring the supply chain. This actually helped this pharmaceutical company to identify vendors with potentially restricted shipments. It also identified countries with higher risk, and then implemented automatic alerts for open purchase orders based on the risk.

It also allowed these pharmaceutical companies’ purchasing function to continuously predict cycle times, and flag delivery time risks, helping it with production planning.

Jon: To take advantage of those use cases that you discussed, a company needs to implement some programs or initiatives, and I’m a big fan of frameworks because they can really jumpstart a project and often provide a proven path to success.

In terms of a framework for building a resilient digital core, I saw Cognizant promoted recently six building blocks to creating a resilient digital core. Could you recap that approach for us and provide some color on the best practices?

Sudhakar: Absolutely, Jon. Me too, I’m also a big fan of frameworks, because Cognizant, all our frameworks are mostly driven by execution led experiences. The one we are talking about here is the six building blocks for Resilient Digital Core. There are six. I’ll probably cover five of those and let me now take the sixth one.

The very first one is to establish a service demand catalog to understand the end-to-end value chain of the processes across all the departments, with a backlog of potential interventions to address bottlenecks through automation.

Including process mining, process simplification, process standardization, and business improvement techniques. That’s the first one, establishing a service demand catalog. The second one is very critical is managing outcomes. There should be a clear understanding of what is success, defined process metrics, such as criteria linked to all the digital initiatives downstream. This I believe some of the companies are not paying significant attention. It’s very critical to make sure the outcomes are managed. 

The third component of this framework is gaining sponsorship from key stakeholders. In this case, we are talking about sponsorship from each layer of management to drive the digital agenda in terms of training of resources, executing projects, and also realizing the outcomes of the benefits that have been established.

As we all know that when we are going through this digital disruption and trying to be resilient, change management is very critical. The fourth one, I would say manage change and develop a very clear communication strategy. Developing these protocols to articulate goals, educate resources, will actually help create enthusiasm across the business on all the stakeholders by showcasing success stories, describing the benefits, the power of their involvement, acknowledging associates, and developing the skill sets.

The next one is about a talent upskill. Upskilling and rescaling the resources is critical. Many IPA tools demand little to no-coding. However, that said, companies need focus teams that include data models, process mining specialists, architects, and developers. A prevailing trend is to improve the skills of process associates to handle simple automation deployment, and these are the five elements of the framework I covered and I’ll let Abhinav touch on the platform and technology.

Abhinav: Yeah. Sure, Sudhakar. Choosing the right platform and technology for building a resilient core is absolutely essential, I would say. Obviously, many companies do it the way they think is right, but what we have seen for experiences is that appointing a steering committee to represent operating businesses, the IT, the CXO stakeholders, and how will you bring an organizer strong vendor ecosystem to create that technology capability for smart orchestration is what it is all about, but how do you choose a platform or a technology?

What I am seeing in the real world right now is that choosing the right Hyper Automation toolkit is basically being followed in such a way that people or customers tend to follow RPA vendors’ toolkit. Many times that is what is happening. Through our experiences, what we believe the right approach should be to first look inside.

What is the entire gap of tools that exist already in your enterprise versus what is available in the market? Do the data gap analysis, identify the differences, and talk to your IT stakeholders. I can use some examples. For example, every organization would have a BPM platform in place. Every organization would have some kind of OCR toolkit that they may have already implemented in their previous life. They may have an analytics or AIML platform to enable through your Data Science Center of Excellence. 

The point I’m trying to make is that don’t just blindly follow an RPA. Just because you have an RPA tool and that vendor has got Hyper Automation platform, it is not necessary that you just choose that platform to be your Hyper Automation platform. It could be well a choice that you make based on what you already have, do a data gap and then figure out how you can integrate and stitch all of these things together. That I think is one of the key learnings from our experience on the ground.

Jon: That’s very helpful and the approach makes a lot of sense. We’ll make sure to provide a link to it in the show notes to the article for anyone interested in exploring the model in more detail.

In terms of rescaling, just one comment, too often we hear that the fear of automation, it’s going to be this massive job killer and major disrupter. I try to advocate a more utopian view of technology, and my hope and belief are really that Hyper Automation and the related technologies really ultimately aid the worker by allowing them to focus on higher-value work and eliminate the more tedious tasks from their daily work activities. That’s probably a topic for another episode. 

To recap today’s conversation with Abhinav: Kolhe and Sudhakar Pemmaraju from Cognizant, certainly one of the largest but most importantly, one of the most admired and innovative companies in the professional services market. We got great insight on how process automation helps businesses manage through times of crises and explore the building blocks to developing this Resilient Digital Core to help companies navigate and ultimately thrive in challenging times.

Thank you Abhinav and Sudhakar for joining me today. I want to give you the opportunity to make a closing comment or provide some final thought, but I also have one final question for you. I’m a bit of an information junkie and always looking for the latest and greatest resources.

My final question for you is what resources, website, newsletter, podcast, email blast, framework, anything at all, do you rely on most to be successful and knowledgeable in your role? Abhinav, why don’t you start us off?

Abhinav: To answer your question, absolutely. I think my current work that I do or in terms of practice development, uses the most amount of knowledge and is on top of things when it comes to Hyper Automation and stuff like that. Obviously, many people follow some of the popular research firms like Gartner’s and HFS from the world, typically, but I also do follow Twitter, a person by the name Cassie Kozyrkov. She’s the head of decision intelligence at Google. I typically follow her for her topics in analytics. The way she talks on analytics, you should just listen to her. You will be just amazed by the clarity of how she brings around topics like statistics versus analytics, and how these topics differ, and stuff like that.

That’s what I typically do, apart from my talking to vendor partners, looking at CB insights, looking at the top AI vendors in the market, doing capability understanding of their presentations, and I happen to be doing the same talk with FortressIQ as well. That’s my two cents on that.

Jon: That’s great, Abinhav. How about you, Sudhakar, any final insights and your go-to resource for staying ahead of the curve?

Sudhakar: Obviously, one of my go-to resources is we now on this team. They are on top of the trend, we follow that. In addition to that, I always stay active on social media a bit, LinkedIn, podcast, or TED Talks. Constantly, I try to dedicate some time to go to social media and all sorts of social media. That’s one.

We also regularly talk to our vendors, partners, and clients to understand what’s happening in the landscape. That’s one more source of data for us. In addition to that Cognizant has a central research team that collects data from multiple sources and provides us the data.

Jon: That’s awesome. For reference, my resource this episode that I can’t miss is MIT’s The Download daily email brief, touches on a lot of current topics around AI, just a great daily read to get a quick check on the industry. We’ll be sure to put a link to all the resources recommended in the show notes. That’s a wrap on today’s show. Thank you, gentlemen, for joining me. Thank you FortressIQ for sponsoring. I’m Jon Knisley and this has been hello, Human.

If you enjoyed this episode, subscribe and check out our series at fortressiq.com/podcast. Thanks for joining us today on hello, Human.