← The Agile Brand with Greg Kihlström®: Expert Mode Marketing Technology, AI, & CX
The Agile Brand with Greg Kihlström®: Expert Mode Marketing Technology, AI, & CX
#713: Agentic AI that improves the customer experience, with Manisha Powar, Qualtrics
00:24:06
Loading summary
Transcript29 lines
- [00:01]
A
The agile brand. Welcome to Season seven of the Agile Brand where we discuss the trends and topics marketing leaders need to know. Stay curious, stay agile and join the top enterprise brands and martech platforms as we explore marketing, technology, AI, E commerce and whatever's next for the Omnichannel customer experience. Together we'll discover what it takes to create an agile brand built for today and tomorrow and built for customers, employees and continued business growth. I'm your host Greg Kilstrom, advising Fortune 1000 brands on martech, AI and marketing operations. The Agile Brand podcast is brought to you by Tech Systems, an industry leader in full stack technology services, talent services and real world application. For more information go to teksystems.com to make sure you always get the latest episodes, please hit subscribe on the app you listen to podcasts on and leave us a rating so others can find us as well. And now onto the show. Agility requires more than just speed. It demands relevance and empathy, especially when AI is stepping in to play a bigger role in the customer experience. What if the problem isn't that AI moves too slowly, but that it moves without context, without empathy and without earning trust? Today we're going to talk about how agentic AI is changing that, offering a way to transform experience management from reactive to proactive and from transactional to genuinely helpful. To help me discuss this topic, I'd like to welcome Manisha Poar, VP Head of Product Customer Experience Suite at Qualtrics. Manisha, welcome to the show.
- [01:45]
B
Well, thank you Greg. Thanks for having me on the show.
- [01:48]
A
Yeah, looking forward to talking about this with you. Definitely agentic, definitely top of mind. But I think the connection here with really, you know, how does it impact and improve the customer experience is key here. Before we dive into all of that though, why don't you start with giving a little background on yourself and your role at Qualtrics for sure.
- [02:10]
B
I started in the software industry 25 years ago and it's really incredible what even in the last 25 years we've come to see and the differences and the step change that we've seen in our industry over the last just last five years even. I started as a software engineer deep deep down in the systems of system area of Windows and then moved into product management at Qualtrics. I lead our product management team for Customer Experience Suite and I really love the bridge that CX product creates between human experiences and technology.
- [02:51]
A
Yeah, love it, love it. Well, yeah, let's got quite a few things to cover here, but I want to start with the zoomed out view of things and then we'll kind of work our way in a little bit. So you've talked about how organizations are overwhelmed by signals but still miss the mark in delivering great experiences. Certainly capturing data isn't the challenge so much, but why do you think that capturing data has gotten not only so much easier but while acting on it in meaningful ways has gotten harder?
- [03:28]
B
That is the central paradox of modern businesses, isn't it? We actually have a saying in the team. Organizations are now data rich, but insight and action poor. We are swimming in so much data, but we are failing to make human connections at scale that that data is actually supposed to enable for us. The capturing of the data has gotten easier because we have so much digital footprint now. Everybody leaves this thing we call signals. We know what's going on. Like you can leave a thumbs up, you can leave a comment, you can talk about the brand on X or other social media channels. You are spending time on your webpage or app like frustrated tone in a voice, in a call center. Signals are everywhere. But really I think the challenge has been threefold. First is just a silo effect. All of this data is fundamentally disconnected. Your website signals and your call center is completely not connected in most cases. How many times do you feel go to the website or the app, you try to do something and then you go call the call center and you have to explain everything again. And in your head you're going like, don't you, don't you talk to each other? Don't you see, like why do I have to explain it to you when I just did it on your website?
- [04:51]
A
That was literally my afternoon by the way. But no joke. Yeah.
- [04:56]
B
So I think that data silos is the first one. The second one is while we, when as we kind of exploded our data signals and our channels and it's a good thing in terms of the amount of data available. It also created so much signal, signal overwhelm because now you are in this place of too much data and by the time somebody creates a report like today, if you're doing this, you're not doing this at scale and humans are doing this work. By the time somebody looks at an insight and acts like decides to act on that insight, life has probably moved on much further. So, so just the amount of signal and the expectation of how fast you need to act on it has just dramatically changed.
- [05:43]
A
Yeah, yeah. And you know, to take, to take that to the next step then even you know, talking about analysis, I mean, so okay, we're you know, we're flooded with data, more, more than we can possibly do anything with. Then we take that and distill that and start analyzing it and then we have all this analysis. But there's only so much analysis that you can do, you know, without action. Right. And so, you know, I, I think even the organizations that get past that, that first hurdle are then stuck with a million dashboards and, and reports and everything like that, which means we need to get to, to action. So I think, you know, agentic and I, I think you've talked about this as well, is, you know, agentic is, is a path forward here. And you know, so first, for those a little less familiar, how do you define agentic AI and how does it kind of fundamentally change the way that some of these challenges can be overcome and to manage these customer experiences?
- [06:44]
B
Yeah, that's a great question. I completely agree. Action is, I think, where the next big frontier is at. The way I look at, I'm sure you'll get many, many different definitions if you ask different leaders. The way I look at agentic AI is it's a system that you can give a complex goal to and it can autonomously reason and understand and then execute on a multi step plan to complete the task. It can make decisions, it knows which tools to use or which systems to reach out to get the job done. And it does all of that somewhat autonomously. Yeah, it fundamentally changes the way action can happen at scale.
- [07:30]
A
Yeah, yeah. And I think this is so important because there's a lot that can be done with if this, then that and the rules based bots and the reactive systems. But we've kind of reached the limits. When we're talking about Omnichannel, we're talking about the segment of want. There's so many buzzwords and terms, but they're all meaningful because the consumer is, you know, they want to access the brand where they are on whatever device and so on and so forth. So you know what separates an AI agent from that chatbot that is very reactive and very, if this, then that kind of programming, you know, maybe share an example of how an agentic system might differ than one of those previous types of methods.
- [08:22]
B
Yes, for sure. So before I jump into a specific example, the analogy I think of is difference between the yesterday's chatbot and what we think now as agentic AI is that between a scripted call center operator and an expert problem solver. Yesterday's chatbot is reactive, it's scripted, it follows a very static decision tree. If you ask it Anything that goes outside of its decision tree, it can't actually, it doesn't understand, it doesn't reason, it doesn't really know anything. It just follows a path. Whereas the AI agent, like a Qualtrics experience agent, is going to be more proactive and goal oriented. It's going to know the context, it's going to have memory, it's going to understand who you are. And that's really where the biggest difference is between a scripted or path based agent versus an AI agent.
- [09:24]
A
Yeah, yeah, I love that analogy there. And for those brands that I'm sure all of this sounds great to everyone listening here and they want to do that. But it's one thing to be, you know, to be proactive versus thinking about being proactive. A lot of, a lot of companies are still, you know, in that reactive kind of like firefighting mode. You know, what does it take for a company to shift from, you know, reacting to anticipating customer needs with, you know, using AI?
- [10:00]
B
Oh, that's a core challenge for so many organizations, Greg. The, the need to shift from firefighting to proactive problem solving isn't just a technology challenge. It's actually a strategic and a cultural challenge as well. So when I think about, or when we've talked to the forward thinking brands and brands that have taken this action or this step, I see kind of three patterns. First one is they look to build a future ready tech and data foundation. Now you can't be proactive if you're blind. So the first thing that has to happen is you have to know how to break down data silos within your organization. You need to know how to unify the data from all channels. You need to have a single source of truth that your agents can actually learn from and leverage. The second one is understanding and setting clear risk ethics and governance policies. With AI and with autonomy, we are going to have a great responsibility to ensure that we have trust. We build trust with our customers. This means setting clear policies around responsible AI use, bias mitigation, privacy protection. And you need a cross functional governance framework that brings together multiple departments, your it, your legal, your business leadership as well as your product leadership. Building that trust is kind of the second core pillar of getting to this jump of proactive agents. And then the third one is, and this is very important, organizations shouldn't try to do everything at the same time. Don't boil the ocean. The best way to build momentum is to identify small and focused pilot areas. Pilot use cases that have extremely clear value that allow you to both validate your Data silo problems or solve your data silo problems but also show you value. And then once you have those wins, you can always scale much more quickly with your executives because you will get a much better sponsorship and cross team collaboration once you start showing the value.
- [12:14]
A
Yeah, yeah. And so to build on that and to keep kind of building towards moving from theory to practice, how do you see, you know, how do you see companies doing this? And you know, in, in the real world with across customer journeys like are there specific industries or maybe use cases where you know, there could be some, whether there's those pilot projects or some first cases or you know, industries already showing promise.
- [12:43]
B
Yeah, the one where we are seeing immediate promise and immediate excitement are high volume industries or complex customer journey industries. Travel and hospitality detail financial services are probably some of the early front runners. For example, in retail you can detect digital frustration during checkout and then an agent can proactively intervene to save your sale, which again in the past you would have seen the sale drop in your post purchase or post cycle analytics and then you would go debug and you would go figure out. But in the meantime you've lost a ton of business. Whereas now the agent can detect these frustrations and course correct them in real time by reducing your cart abandonment. Imagine that an AI agent notices you return three pairs of shoes of the same size. Instead of just helping you process another return, it could ask like are you having trouble with the fit? So you can imagine this is also just an incredible product improvement opportunity. Similarly, in travel, if your flight was canceled and there is a bunch of pressure that you're going through, you could just quickly have an agent help you with a bunch of things that a human can also do. But many of these underlying tasks are pretty automatable like book your car, set up your next travel booking or set up find a flight and change your car reservation. So you can imagine an agent being able to fulfill that, an AI agent able to fulfill that really, really fast without having to wait in a queue for an hour for a pretty high volume use case.
- [14:37]
A
You know, I think a lot of the use cases that, that may come to mind first for a lot of people, a lot of organizations and teams that are thinking about agentic are technical ones or efficiency gains productivity. But you know, this idea of empathy at scale, you know, it's a, it's a pretty interesting concept especially when we're talking about, you know, we're talking about AI agents and, and empathy. So what role does customer context, whether that's survey data Reviews, call transcripts, and even interactions on an app or a website. How do all of these help agents actually act with an understanding of customers?
- [15:19]
B
Yeah, I strongly believe that customer context is everything. Without it, empathy at scale is just a buzzword. You can't empathize if you don't understand. And the way you understand is by listening to what's happening, what your customers, what your users are currently going through. So for me, these experience signals you talked about are really just that fundamental building block for a truly empathetic AI agent, whether it's a human agent or AI agent, actually. But in the world of agentic AI, it becomes even more important because now you are relying on this AI agent to act on your brand's behalf at scale. And this also goes back to the data silo problems we were talking about. If the AI agent is only seeing that you gave a low score on a survey, but it has no other context, it's not going to really be able to help your customer that much versus if it had everything that this customer has done on your website and your app and your call center. But now that agent, the agent, has a lot more context and a lot more memory about who you are and what you, what you, the human that, what you, Greg, are going to need and how to personalize the actions just for you.
- [16:35]
A
Yeah, yeah. And so then, you know, tying those two things together so that, you know that, that empathy or even, you know, authenticity as well as that need to drive productivity and efficiency, there's a, there's maybe a little tension even there. You know, I know there's always a lot of, especially of late. There's a lot of emphasis on, you know, how do we save dollars, how do we, you know, how do we do more with less and all that. And yet, you know what you're saying, the context and the, the empathy is what also really creates long term, loyal, lasting relationships. How do organizations balance this? You know what, what should leaders be thinking about to balance these two things? And can agentic AI really deliver on both?
- [17:26]
B
Well, thanks for asking. What I think is the most important question, Greg, this is where businesses really want to help customers, but this is where the tension happens. Because businesses want efficiency, they want to save money, they want to make money. But customers don't want to be treated like a number. Nobody wants to be optimized. They all want to be understood. But at the same time, the business leaders don't wake up every morning and say, I'm going to treat, I'm excited to treat my customers as Numbers Today they want to provide a personalized and really human experience to their customers. They just couldn't do it at scale. And that's really what gets me really excited about agentic AI because when it's designed correctly and it's given the right data and the right autonomy, it can actually absolutely do both. It can give the business leaders the efficiency, but at the same time gives the customers the experience that they deserve. It can augment your human agents and it can take away mundane and repetitive tasks that nobody really wants to be doing as a human agent in the first place. So I can handle all of that scale and it frees up the human agent, your human workforce then to do truly high value work.
- [18:48]
A
Yeah.
- [18:48]
B
And also gives them the superpowers of personalization.
- [18:52]
A
Yeah, absolutely. And I think, you know, the getting to the first step of, you know, being able to, you know, get, get past the signals and moving towards action is, is a great step. And then, you know, then there's sort of beyond recreating what's there. There's so many new possibilities as well. Right. So I wonder, you know, what kind of skills or even mindsets are going to be critical for leaders that are building the next generation of, you know, not only experienced platforms, but those on their teams that they're going to come up with stuff that we haven't even thought of yet. Right. Because they have tools to, to do that, you know, what's, what's, what's going to be crit in the months to come.
- [19:39]
B
That's a great question. And I think the technology with AI is evolving so quickly that a lot of this is going to change as we speak, even as we get through the next couple of years. But I do think the leadership mindset also has to evolve and stay open to this. So I really see three critical parts as an executive. The first one is shifting the organization from departmental thinking to a journey oriented thinking. Going back to what we talked about. Your customers don't look at you as a department, they look at you as an organization, they look at you as a brand. So you should no longer be tolerant. Tolerant of the this is marketing team's data or this is the support team's problem. Because your agent AI needs to operate across all of that. So that's the first one, just shifting your organization from departmental to journey oriented thinking. I would say the second big one is becoming ethical executive in the organization. As AI becomes autonomous, leaders must become the arbiters of ethical boundaries. This is not just about can we do this? The business Leaders also need to ask about should we do this? And if yes, how. And then the last one is what we started with, which is they have to embrace an agile and experimental mindset. We are in such early innings of this transformation that having a static, static mindset is not going to be helpful for our leaders. So start with high impact pilots, learn quickly from success, celebrate the failures, try the next pilots, next experiment and create that culture not just for yourself, but also for your company.
- [21:28]
A
Yeah.
- [21:29]
B
The future is not just about technology expertise. It's also about creating that customer centricity and ethical governance and agility in the organization.
- [21:38]
A
Yeah. Yeah. Love it. Well Manisha, thanks so much for sharing all your ideas and insights today. One last question for you before we wrap up. What do you do to stay agile in your role and how do you find a way to do it consistently?
- [21:53]
B
Part of it is you and I met at a conference just not more than a day ago. Part of it is really staying open to things that are happening in the industry. I have a lot of conversations with customers. I have a lot of conversations with leaders in this area. We have Gurdeep who leads our he's a visionary AI leader. We have had conversations with a bunch of software vendors that are now going far beyond what originally anybody could thought possible. So just staying open, leading a lot, listening to a lot of lot of podcasts, a lot of new things that are happening and creating that. Same as what I talked to you about. I have the responsibility to do that in my team as well. So we experiment with AI in our team. As a product manager, how are you going to be a better product manager because AI is now at your fingertips. So there is a lot of experimentation that we do within our organization as well. So. And we have fun with it.
- [22:55]
A
Yeah. Love it. Well, again I'd like to thank Manisha Powar, VP Head of Product Customer Experience Suite at Qualtrics for joining the show. You can learn more about Manisha and Qualtrics by following the links in the show notes. Thanks again for listening to the Agile brand brought to you by Tech Systems. If you enjoyed the show, please take a minute to subscribe and leave us a rating so that others can find the show as well. You can access more episodes of the show@theagile brand.com that's theagile brand.com and contact me. If you're interested in consulting or advisory services or are looking for a speaker for your next event, go to www.greggkillstrom.com that's G-R E G K-I H L S T R O M.com the Agile brand is produced by Missing Link, a Latina owned, strategy driven, creatively fueled production co op. From ideation to creation, they craft human connections through intelligent, engaging and informative content. Until next time, stay curious and stay agile. The agile brand.