The Global Startup Ecosystem: Reimagining Business Processes

Jeff Kavanaugh: Welcome to the Knowledge Institute Podcast, where we discuss the startup ecosystem with experts, deconstruct main ideas, and share their insights. I'm Jeff Kavanaugh, Head of the Infosys Knowledge Institute. And today, we're here with Satish Nair, Vice President of Infosys Business Process Management. Satish, thank you for joining us.

Satish Nair: Thanks for having me over, Jeff.

Jeff Kavanaugh: Satish, you've been vital to the strategy and development of Infosys BPM and its work with startups. Tell us a bit more about your role in leading the Infosys BPM practice.

Satish Nair: Digital business services within Infosys BPM delivers the strategy of the digital pentagon, right? Driving the experience, the effectiveness and efficiency metrics for our customers. We do three things. Number one is the exciting part of it, which is truly developing and thinking through what are the nature of solutions to solve business problems, right? Bringing in the latest and greatest of what's happening in the digital space. Second, it's about how do we really sell it through the deal? How does it really enable you to win the deal? And the third part of it, which is the critical part, is how do you land it on the ground? So it's not just about defining the solution and taking it to the customer, but it's also about delivering it to the customers and showing that value.

Jeff Kavanaugh: Well, that's interesting because I think most people will focus on the delivering the value piece, but there's a lot that goes before that as well. Let's dive in to one dimension of delivering the value because to the uninformed observer, BPM, business process management, might appear to be a kind of a mature, dare I say, boring area. What is it about business process management that's so vital to solving your clients’ problems and where do you see it going?

Satish Nair: It's interesting to say that Jeff. The one thing that is actually changing in business is the business itself. And what that does is bring fundamental changes in every part of the business process that you do. Be it enterprise functions like sourcing, procurement or the management, your finance, accounting, HR. Or your core industry processes like mortgages, or underwriting the insurance work. All of these are changing rapidly. What people don't realize is the core of digital that's actually getting adopted at these industries. Especially if you look at within the B2B segment, too, is, fundamentally changing the way operations or the processes itself it delivers. What it also adds is the confluence of all of these changes, be it in terms of data and everything that's coming from within the data side, what's happening from a technology, be it adoption of machine learning, AI. Right? What's happening from digital new, digital experiences, new channels of the way we actually deliver comes to reality at the business process end, where it hits each of the consumers and the people who actually deliver these services across.

Jeff Kavanaugh: Well, if business is changing so much then you think it'd be hard to keep up. And I know one aspect of our discussion's on startups, why are startups so important in this decade of exponential change and in business process management overall?

Satish Nair: By default startups have the inherent DNA to focus on the best of what technology offers today and attracts the best talent. And most importantly brings in the concept of agility and speed and the very hunger that they show to come and solve a business problem. So bringing together the demands of a changing business and the three areas that I believe startups have a lot to offer means a compelling solution that we can take to solve that business problem.

Jeff Kavanaugh: Well, let me try to make it real here. Can you give an example of a startup you've worked with recently and the problem it solved, maybe that distinct aspect of being a startup and how it made the difference?

Satish Nair: We've been working on a very specific area within mortgage in the post-closing space where we get these 100 plus types of documents that were coming in. All of this had to be classified, key elements of data extracted in a very limited amount of time. And what happened during that period was, we approached large players, established players, technology that we think we can pull together from generating AI that was available out there and into the mix, we introduced a small startup who just got initial funding and come up with a very interesting AI led solution in the market space. It started off as a small prototype or proof of concept across all three. It was an interesting challenge that was given to all three and said, "Here you have the set of information. How do you really come up and craft the solution?"

Satish Nair:

The first one was a large established player who wanted all the constraints, all the details of what was required and came back with a timeline of, "Hey, listen, we'll come back to you with a proposal within two weeks. And by the way, we think we'll be able to use AI and extract first pass 60% to 70% of the information that is required for this process." The startup on the other hand started with, "Can you give me the data?" And independently comes back to us and says, "Can you give us one week with that data?" And guess what happened?

Satish Nair: At the end of the week, they came back and she said, "Listen, we've done a first pass with the data. We think we can get only to about 80% throughput, through the AI and machine learning that we've actually built specifically for mortgage documents." Fast forward to two weeks later when we got the proposal and what we think from all parties, the final output that was given, or the final expectation that was provided for the three players was "Yes, you have another 10 days for you to come back and give what the outcome of your prototypes are." The large player came back with saying, "Hey, listen, out of the 100 plus document types, we were able to train 40 of them. And we think we'll get to an accuracy of 80% to 85% on those 40 document types."

Satish Nair: The startup on the other hand came back and told us, not just told us, showed us that, "Hey, listen out of this 140 document types or a 100 plus document types, they're there. I think we couldn't do about 10 of them. We're sorry. But by the way, we were able to increase the extraction rates on training of those documents automatically. And here's the outcome, we think we were able to get to 90% and by the way, if we get this deal, we think we'll go upwards of 90% extraction." Now think about what happened. Number one, the startups are very, very focused on the mortgage industry. They were very focused on the mortgage document type, helping them to deliver more than 90% of the document throughput because they understood the domain. Understanding of the domain also meant that the time required to train them or give them any sort of knowledge was bare minimum.

Satish Nair: Second, if I look at the process, these people and the approach that they took to… “Forget about the bureaucracy, tell me the document, and this is what I will do with it." The sheer speed that they brought on to the table is absolutely fantastic. And by the way, from a technology standpoint, which you would think that in today's playing field, a larger organization might have a lot more depth. Interestingly, both were on similar open source stacks, right? So in technology, they were at par, the focus on business plan far ahead, agility and speed meant that the time to value was, just in the pursuit stage, was probably an 80% or 90% better. Interestingly, obviously we went ahead with the startup could deliver the program. That interesting part was the program was delivered ahead of time at far, far better quality than came out during the proposal stage. And more importantly, at a fraction of a cost that you would expect in a program of this nature.

Jeff Kavanaugh: To summarize what I heard is data agility, speed, and depth, you're swapping one for the other. And it's a very compelling argument to introduce startups into what traditionally has probably been more of these established players, trying to make sure large companies get large providers. Can you give us an overview? This is one example. Broadly speaking about how you engage startups. So in that case, they did a prototype and then give you quick answer. Are there some broad categories or stages where you bring them in? And what's the role that a startup plays?

Satish Nair: So in learnings over time, that different startups, we tend to break it down into three parts. And this is where keeping in mind that all is not hunky-dory when it comes to engaging and bringing together on one end large organizations, on the other end, small nimble startups with sometimes not as much deep pockets. So bringing these two together meant a lot of learning for us as we've been through this program. The first and foremost was to set expectations with both parties. And that expectation setting also meant that we had to dirty our hand and try it out, try to see what the challenges are in engaging with these startups. Setting expectations on areas like risk, compliance, scalability, the in-depth testing that you need to do on some of these, technology to scale out became very important. So the first phase was literally setting up a platform for the test outs, at least the bare outlines. Simple things like, will it pass the information security departments of clients?

Satish Nair: Will it pass scalability testing? And remember these startups are coming in and did not have most of these chances to test out some of these things at scale. So, that was the first sandbox environment that we played on them. The second part of it is as we go through the sales process or the engagement process with the customer, it's about making sure that startups understand that we large corporations buy, large corporations engage. What are the commercials, right? And how do you come up with a win-win situation for both? And the third phase of it was on the implementation side, where it's about doing the pilot, it's about scaling it out, but more and more, we see a third element of it.

Satish Nair: And this is where the hunger of startups really come across, which is how do you add completely new capabilities to the existing base, use case that they come up with? Because a lot of these ideas that come up from startups when they focus on a specific business problem, it is very easy once they establish to be able to scale it out, into adjacent capability, a new muscle that the startup can also build and learn that adds value to the customer.

Jeff Kavanaugh: Well, to add to our keyword, adjacency, I think you added two more. One is intersection. You're seeing how these different processes or domains come together and then finally, convergence. Maybe that's one of the most important reasons after all, as these different industries are converging, so they're able to connect the dots and maybe that first problem gets solved. And at the same time, it leads to a natural sequence of others. Have you seen where solving Problem A, the thing that got you together, led to solving a very different and maybe more powerful problem in the future? Other words, you went from the initial problem and then it naturally led to a second or third one?

Satish Nair: If you look at, say, document extraction in financial services are the key, for example, in solving a particular problem. Or you look at risk management and supply chain as a core issue that got solved. The key advantage that some of the startups are able to show is the agility in pivoting onto a new opportunity, right? And to enable this pivot, they don't have a problem or no ego in working with experts in the same industry. And let me give you this example, when the solution around LightBulb, in the context of Europe had to be extended to Down Under, the team at the startup, all they had to do was call up partner, engage with the domain, consulting one of the domain consulting folks, some small organization based down under. The whole process that would have taken ages in any large organization, going through conversations and embassies and NDAs and commercial discussions and everything else.

Satish Nair: These guys were able to pick up the call, said, "This is what we are working on. This is the objective of what we're trying to do. We're going to take this solution. We think there is an opportunity Down Under, and we need you to work with us." And within a week you had the first discussions going on about, how does risk differ in different markets? Looking at it from risk and compliance. Now that agility that startups provide in being able to attract partner with specialists, without ego, without quote-unquote a lot of bureaucracy, that I think adds a lot of value in terms of how quickly they adapt to adjacent opportunities.

Jeff Kavanaugh: What are the three things you'd like executives or listeners to take away, as you think about startups and the BPM ecosystem?

Satish Nair: One of the feedback that we got from one of the startups actually put things in perspective. They said, "You know what, the fact that you were completely transparent about the opportunity, got us involved on day one, talks a lot about the values and ethics of what you bring onto the table. But more importantly allows us to put in the best effort that we can to solve the problem." Right? So as providers, I think it is critical that we are transparent, bringing together the large customers and the access to customer base with the startup ecosystem. The second part is to be completely transparent with the nature of problems that you're likely to solve. From a customer standpoint, it is about making sure that you understand the way to engage a customer, the fact that they need decision-making speed to be a lot faster.

Satish Nair: They don't have that depth of pockets to wait around, waiting for multiple decisions to happen. You're aware of the base foundational work around providing the sandbox environment, providing that the decision making for things to move along. And from a startup standpoint, it's also the ability for the startup to be razor sharp focused on business problems. Like you mentioned, Jeff, on the opportunity that'll arise from adjacencies, and be on top of your game from a technology innovation standpoint. So as players like us, our ability to bring together and solve some of these “flow problem” across large ecosystems of customers and the nimble, agile startup ecosystems, I think plays a huge role in making this a win, win or rather win, win, win for all parties involved.

Jeff Kavanaugh: As we wrap up Satish, what resources do you recommend so people can learn more?

Satish Nair: Please take time to read through the work that is being done within the IIN Ecosystem, the Infosys Innovation Network, and you can see it on the Infosys website. You should also definitely read up on some very interesting articles out there, including one that I read recently on HBR, on the challenges and the opportunity that startups actually provide or how do you really look at understanding or being forewarned about how to approach some of these areas.

Jeff Kavanaugh: For these resources and others, as well as to reach Satish, you can find details on our show notes and transcripts at infosys.com/IKI in our podcast section. Satish, thank you very much for your time and a thought provoking conversation.

Satish Nair: Thanks a lot, Jeff. Thanks for having me out here.

Jeff Kavanaugh: Everyone, you've been listening to the Knowledge Institute, where we talk with experts on the global startup ecosystem, deconstruct main ideas and share their insights. Thanks to our producers, Catherine Burdette, Christine Calhoun, and Dylan Cosper.

Until next time, keep learning and keep sharing.