International Business Machines Corporation (IBM) Earnings Call Transcript & Summary

Once we bring this altogether, AI and digitally connected services are the key to transform customer support, optimizing the capabilities of smart machines and even smarter humans. But let's talk a little bit about how that works to deliver more rewarding and personalized experiences.

Well, the combination of AI plus human agents can deliver a level of responsiveness that is impossible with humans alone. Take AI-driven cognitive assistance, which use natural language processing to engage with consumers using real conversation. These emotionally intelligent systems understand what people say as well as what they mean and even detect sentiment and context to respond in meaningful ways. So they can offload routine inquiries to offer instantaneous answers, connect people to the right information, entertain, educate and even help them complete health-related tasks.

Absolutely. They create entirely new way to engage while eliminating the complexity of keyboards and touchscreens. So a customer can be guided through interaction with a cognitive assistance that can communicate with natural language to both simplify and expand self-service opportunities through whichever channels the customer prefers, whether it be messaging, mobile, web, effectively working with me in the manner that I choose.

But AI can also provide live agents with insights to solve complex problems more quickly. For example, I may reach out to a contact center website, but the information I need is nuanced. So I may get part of what I need, but then they want to move on to a live agent.

In that scenario, AI can assist in a number of ways: first, by engaging with the customer to meet their needs as much as possible; then managing a seamless handoff from virtual to the live agent and providing the information already shared by the customer so that they don't need to repeat their process. But it can go further, analyzing in real time, customer data as well as past interactions to provide knowledge to the agent to better understand that customer and their preferences, broader health interest, even goals or motivations. Combining more data from more sources, this kind of AI assist allows human agents to solve problems more quickly, and importantly, to offer more valuable support and personalized attention.

And with 5G rolling out, bringing blazing data speed, virtually no lag time, the pace of digital innovation will accelerate. We're already combining mobile, AI, natural language processing, IoT and augmented reality to create more interactive, multidimensional experiences to educate, motivate and support new health services. And as an example, we recently worked with the children's hospital to create an augmented reality mobile-enabled app to help relieve anxieties for young patients. An animated avatar on the app serves as a companion as the child moves to the hospital and procedures, sharing information, entertainment, games and rewards.

Well, we want to allow enough time for some questions. So let's wrap up by recapping the main takeaways. AI-human collaboration offers a way to elevate customer services, combining virtual care with intelligence-driven capabilities and automation to create personalized, efficient on-demand experiences. Considering all the ways digitization has transformed how people shop, travel, bank, we've yet to see significant change in how we deliver health care services. Yet now, more than ever, consumers are seeking digital-enabled support, trusted information, and they want 24/7 help to meet their in-the-moment needs. Today's disruption creates an imperative to prepare for the new normal, with increasing need to find more capacity to engage consumers across digital and physical, and as well, empower a collaborative, interconnected care ecosystem that can transform individual experiences and overall health.

All right. Great. So we've had some good questions come through during the presentation. [Operator Instructions] All right. So starting with the first question that we have in here. Let's see. Okay. So you talked about building a road map, end-to-end customer services and looking for quick wins, what might that include? All right. So great question. Probably won't be a surprise that my answer is it depends. A very fast answer, at least, and really that's because it's all driven by empathy for users. So one -- and it starts with really getting that deep understanding, collecting it together, driving out an end-to-end vision and then kind of working backwards from what capabilities are we starting with and what can be quickly enabled against that future state vision. I'll give you a couple of examples. So one organization I worked with, they were getting absolutely inundated with questions about password reset, among all sorts of other things, and particularly, we saw spikes in new types of questions coming in with COVID-19. It was a major source of frustration to members that they weren't able to get their password reset very quickly. They had to call in. They had to wait. And the reality was the capability actually already existed for self-service password reset, but it was a simple -- relatively simple, moving how -- where that existed in the flow and the experience online and mobile so that members were able to more quickly do that and didn't actually have to go through the painstaking process of searching around to find the ability to do it in self-service or calling in. So that's one example. The reality is there are a lot of different things and starting points in terms of what qualifies as quick wins. And one interesting thing that we recently did was a crowd-sourcing exercise at a hospital system where we actually -- we polled the audience for what were some of the places that they saw that were right for human center design, changing the experience around customer service. And it included a number of different things, but it included proactive outreach for denied claims, personalized education and information to support post-care and even pushing contact center-only processes to mobile or web channels. So those are some different ideas around what might qualify as a quick win while recognizing that it all has to build back up and ladder up to that future state vision of what do you really want to do, who you want to be, what's your brand promise. All right. So with that, let's see. The next question I see in here, "Can I get knowledge bases and AI assistance up and running quickly? Or does it require a long time to train with the right information." So Deiva, do you want to go ahead and answer that one?

Sure, Melissa. So I think there's 2 parts to that question as I'm looking at it as well. Can we get knowledge bases and AI assistance up and running quickly? And then does it require a long time to train? So a great question from a speed-to-value standpoint. The short answer is yes, we have AI assistance up and running in 48 hours to support several use cases of business functions. So it's a pretty automated enablement process, if you will. And with respect to additional training or what we call as intent, which is a conversational concept around the customers' intent, whether they're asking a question, they're trying to find information, understanding the true intent, irrespective of the natural language interaction itself, the words themselves. So with respect to additional training or configuring additional intent to support their various needs from a customer perspective, we enable that using an agile approach with very rapid deployment times, even as frequent as once or twice a week, we're doing some of this work in the COVID-19 space where speed is of the essence when it comes to keep finding information in patients, members of health plans, trying to find information about testing and treatments and so on. So we're able to do that with an initial set of intents to drive speed to value in 48 hours and then additional training or intents using an agile approach based on the customers' needs. Back to you, Melissa.

Fantastic. Thank you so much Deiva. Okay. Next question I see, "So to introduce AI virtual conversational assistance, is there any -- is any of that technology preconfigured to save time on implementation?" Clark, do you want to get us started with that one?

Sure. You know what, I'd like to hear from Deiva first and see how he approaches this, and then I can probably follow on.

Perfect.

Sure, Clark. Yes. I think this is similar to the previous question in terms of is any of that technology preconfigured, so the answer is yes. And that is the case across several use cases. I'll give a quick example. As I mentioned, for COVID-19, working with health care payers and providers, we have Watson AI assistance up and running in actually to less 48 hours. So what's enabled initially in these cases is an initial set of COVID-19 specific intents, the types of questions that whether you're looking from a provider perspective, what patients would have us questions around testing, testing locations, and it is all based on the CDC guidance as accelerated to shorten the implementation times. But in 48 hours, we have initial set of intents deployed rapidly to serve the needs of customers, and we've also added intents in an agile manner. And this is working closely with the call centers and trying to deflect those calls because we don't want the call centers to be burdened -- overburdened, I should say, and the AI assistance can certainly learn pretty quickly, but we are taking insights from the call centers on the types of calls that we're getting and rapidly using an agile approach, enabling -- using Watson AI assistance to [indiscernible] questions that members and providers and patients may have. And this is to drive the best possible experience despite the dynamic conditions as a result of the pandemic so that customers are not waiting on the line trying to speak to someone, or they're able to get the right information quickly for them to address the current situation.

Yes. Thanks. I thought that's perfect. I think yes -- no. I think the piece of the puzzle that I look at is the amount of time and capital that's been invested into the cognitive platforms to allow for open APIs to be leveraged from a simplicity of tying them to a platform, integrating and implementing them into the platform. But the thing that you touched on [indiscernible] critical is that the assistants are learning over time, right? So they come -- we can have the AI tool set show up with intents that are preconfigured. But as it learns, it becomes more robust. And with that, it becomes more of an assistant not only for the client and the customer, but for the employee as well. And that's where we see a lot of our clients investing most is actually in the employee and experience because if you increase the employee experience and you make it easy for them to do their jobs, then the AI tool just becomes an enhancement and an extension of that relationship between the internal brand advocate and the external client customer.

Yes. That's an excellent point.

Awesome. Thank you, both. Okay. So next question in here, "With the platform approach for customer services, how easily can that be customized and linked to other customer-related tools for CRM, marketing or other care services?" Clark, why don't you take that one?

Sure. Sure. I mean, I think it's -- a platform approach is paramount, right? And there are CRMs that are in the marketplace. Obviously, I know Salesforce have been working on that platform now for 8 years. You see that platform grow and morph over time, right, whether it be from a sales or a service or marketing, but organizations need to understand the customer journey. They want to understand the relationship between the brand and them. And without that ability to make it extensible, you lose out on critical touch points. We recently worked with a large state who's facing the backlog in the amount of COVID. And they came back to us with a pretty simple use case of reporting capabilities. We came back to them with an online tool set that was built on the platform, pushed data into service cloud, in this case, and then was able to move that data up into the governors' offices to allow for real-time decision-making. But the ability to heat map and understand where there are critical areas of need and critical areas of care avoidance allows our governments and our state and local governments, for that matter, to become well informed with data that they would not normally get out of the existing archaic technology that they have. But they see this now as a tool set that can move them into delivering direct care between patient and provider, understanding what resources providers are lacking. And in a pandemic like this, if you can't make those decisions in a real-time basis, then people will suffer. I hope that answers that question.

That's a great answer. Thanks, Clark. Okay. So next question, "What kind of use cases do you see for mobile and 5G delivering in customer care?" Another fantastic question. I'll start off and then invite Deiva and Clark to chime in as well. The first thing that comes to my mind is actually the case study, Clark, that you alluded to, right? So an avatar that actually in a pediatric setting that really is helping children through the entire hospital and care journey from making -- keeping them engaged, providing some emotional support, right, as they're going through treatment. I think that's a really fantastic use case that speaks to the power of where we're going with emerging technology. And then a couple other things that come to mind, way finding in a hospital setting, certainly, is the abilities around that are amplified with 5G and mobile. I even think about physical therapy use cases, post-discharge support as another place where there's actually an incredibly powerful ability to change the entire experience and the way that individuals are receiving a more holistic care through the application of 5G and mobile. Clark, Deiva, anything that you'd add? Any other use cases that come immediately to mind?

I think the explosion of -- first of all, we're a mobile society. So if we're not engaging through mobile device, then you're probably missing a huge mark in your ability to connect with your clients and your customers, and in this case, patients, right, is that what I think of immediately with mobile and 5G is this explosion of telehealth. And the providers that I've talked to see telehealth as a whole new means to get to patients who now are homebound. But some were homebound months ago. So people who live in areas that don't have access to help immediately or they're afraid because they live in depressed neighborhoods to actually leave their homes to get to the doctor. Telehealth is going to open all kinds of possibilities for better care and showing kind of empathy and compassion and give people what they need, when they need it through networks that are widely available.

Yes. That's a great point. And I was going to give an example around telehealth as well. We're working with a provider -- health care provider system in Massachusetts where, because of the current situation, the vast majority of the patients are disengaged. And we are helping them, and as part of the overall engagement strategy, 5G-enabled devices that can help monitor vital signs. So that's helping providers to provide the continuity of care. And I think that's just one example of several known examples or use cases as the question had asked here. But there's lot of use cases and capabilities that don't exist today that will come into being tomorrow. And the more we understand the customer and create that 360-degree view, if you will, the better customer service we can provide. And I think the promise of 5G in terms of building that understanding around the customer, holistically speaking, and the promise of AI to make sense out of it to provide better customer service, I think that's -- those are the use cases in front of us that would benefit our customers, both in health care and life sciences. Melissa, back to you.

Yes. Fantastic. It's a great answer. And so that actually does wrap up. I don't see any other questions in here. So with that, we'll go ahead and wrap this up. But I want to thank everybody so much for joining today's webcast. I really appreciate Clark and Deiva for taking the time to join me on this topic. It's clearly one that we have a lot of passion around. And even more so, I want to thank you -- thanks to our audience because we know there are a lot of competing demands facing this community in particular. And the work that this team is doing is more important -- is so critically important to what our nation and our world is facing right now. We're actually incredibly excited by the opportunity and the ingenuity that we're seeing sail out in the market, in particular in health care and life sciences because organizations are really sprinting to redefine customer service. I hope that we've given you some ideas and perspective on how we're thinking about this space and really look forward to continuing the engagement. So encourage you to download the materials, reach out. You've got our information. And really, thank you again for joining today's dialogue. With that, we will wrap this discussion up. Thanks so much.