← The Agile Brand with Greg Kihlström®: Expert Mode Marketing Technology, AI, & CX
The Agile Brand with Greg Kihlström®: Expert Mode Marketing Technology, AI, & CX
#792: Replay: Increasing speed to insights with Sid Banerjee, Medallia
00:29:48
Loading summary
Transcript37 lines
- [00:01]
A
The agile brand.
- [00:10]
B
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. Happy holidays from me and the rest of the Agile Brand team. Just a note before we get started. This episode ran earlier on the show this past year, but I wanted to share it again because it has some really valuable insights in it. I hope you enjoy what could your CX teams do to strategically move your brand forward if they weren't tethered to dashboards? Agility requires CX teams to move beyond reactive reporting and embrace proactive insight delivery. In an era where insights are instantly available through conversational AI, the strategic role of CX is about to shift big time. Today we're going to talk about how AI is reshaping customer experience by freeing CX teams from dashboards, static reports and manual data analysis and allowing them to lead strategically with real time intelligence. To help me discuss this topic, I'd like to welcome Sid Banerjee, Chief Strategy Officer at Medallia. Sid, welcome to the show.
- [02:01]
A
Thank you Greg. Great to be here. Yeah.
- [02:04]
B
Looking forward to talking about this with you. Before we dive in though, why don't you give a little background on yourself and your role at Medallia?
- [02:11]
A
Sure. Happy to do that. So I've been here at Medallion now for about six months, actually almost seven months. Started at the beginning of the year in a Chief Strategy Officer role. My primary focus is really helping to advance our market position, make sure that the market knows about all the great innovations that we're developing around AI and around automating and operationalizing customer experience. But I'm also working with a lot of our customers and a lot of the teams that support our customers to make sure that we're really evolving the way we think about cx. A lot of the things we're going to be talking about today, actually.
- [02:45]
B
Yeah, love it. So yeah, let's dive in and I want to start. We're going to talk about a few things here, but I want to start with the shift in mindset required for CX leaders. And certainly the idea of the data driven leader and the data driven company is certainly top of mind. You've said that we're moving into a post dashboard era though, and that's a pretty big statement. Lots of people probably have a dashboard up on a screen somewhere even as we speak. But can you explain what does that mean for how CX teams are operating today?
- [03:18]
A
Yeah, I think I've been definitely on this soapbox for a few months now. And yeah, I can give you maybe two perspectives. So first of all is that when you think about how CX has evolved over the last 10, 15 years, a lot of the companies that help enable customer experience improvements have done so in a very kind of one way approach to getting information about the customer into the hands of people that can take action on it. And it's really about getting data into a system, analyzing the data and then exposing that data through dashboards. Right. That show you why are my customers calling? What are they complaining about? What do they like or dislike about a product or service? Are they likely to recommend that product or service to their friends based on the experience? And it's very much driven by analytics that are visualized through a computer screen like the ones we're looking at right now. And it's a very one way process all the action. All the intelligence relies on the human to basically process it in the human brain and then to take the action with the arms and legs of the human. And that's a perfectly good way to do it. But when you think about AI and you think about the technology innovations that have happened over the last 10 years or so, you can draw a line to a world where that's not going to be the way things happen. And I like to tell the story of when I first got my Alexa right and most of us probably have Siri or Google Assistant or Alexa. We start figuring out that we can ask it questions and it gives us answers, it doesn't just give us data. As the technology gets better, it's actually better at even solving problems or recommending things that you should do. And all of us have some exposure to using these dashboard less screen less interfaces in our personal lives. And I think more and more of us will even over the next few years with ChatGPT and other things, but we still look at scorecards and dashboards at work, right? There's a world where if you can take this information that lives in CX programs and in CX systems and CX dashboards, upload them to a system that is a true assistant, that can actually look at the operational inputs and outcomes, that can actually discern trends and patterns from language and emotions and sentiments, and then actually process that information like the human brain does, and then talk to the human in a conversational way. Very quickly. If I'm a store manager, I'm not going to look at a dashboard to figure out how I'm doing. I'm going to ask my CX assistant what do I need to focus on today. Or if I'm a product manager and I want to know the most important issues that are driving product returns or that are driving product questions after a marketing campaign, I will ask my assistant and very quickly that assistant will be my coach, it'll be my enabler, it'll be my business process improver, it'll help me to make decisions without having to stare at dashboards. I think where we're ultimately going is that those assistants won't just tell you what you should do as a human, but where they have confidence that they can actually take the action themselves, they will do that. And that's when AI becomes, to use a very kind of au courant word right now, but they become agentic, they take agency and they will actually close that loop for you. And we just all need to think about the paradigm of computing that we've all grown up with and lived with over the last 20, 30 years and recognize that it is changing and it's going to change not just for consumer use cases, but for the customer experience, professional as well as.
- [06:38]
B
Yeah, definitely. And you know, and that paradigm is, you know, a few things, I mean, means that we're often reactive a little too late, you know, because we're, we're spending our time staring at dashboards and pie charts, wondering what that means and, and trying to interpret it with sometimes only a single view of, of that information and not the fuller information and, and the real time nature of some of this data and how AI can respond makes it valuable as well. So, you know, what is, what does that mean then for the role of the CX leader? You know, how does, how does that evolve? And I would maybe even say elevate, you know, by, by using AI in this way?
- [07:18]
A
Yeah, I think, well, there's a few things. And again, this is Going to, I think, rapidly evolve and change as the technology becomes more capable, but at a minimum. And I'll kind of go from kind of most tactical to maybe most kind of most revolutionary revolutionary. As AI gives us suggestions and predictions and can help us get faster from insight to action, the human will do that last mile of decision making and not frankly, verification, right? Because I think there's still going to be a period of time where we get used to the new technology. So the CX professional may spend less time looking for the insight and more time validating the insight and the recommendation. Think of it as last mile decision making. The second thing is, as the AI gets better at actually taking actions, we will often still want to have a human in the loop to say, yeah, I sign off on that action. So think of it as human verification of actions, not just of insights. So agentic won't automatically take over our jobs. It's going to still need that human in the loop to make sure that it's the right thing to do. And ironically, that will then train the AI to get better and eventually they will probably take more actions over time. So let's imagine a world where now the AI is doing the insighting and doing the actioning. What's our job? Right? Our job is still to make sure that the AI is learning and evolving. Because one of the things I've learned as I think about AI today is the models are never perfect and they're never complete because the world is changing, right? There's new information, there's new products, there's new news, there's new world leaders. AI has to constantly be informed about what's new that is not in its model today. If it has to infer everything, it's going to make mistakes, it's going to hallucinate, right? So our job as professionals will be to make sure that the AI is as current as it needs to be to be as useful as we want it to be. And think of us as the tuning and evolving capabilities. We're all going to have to learn new skills about how to teach our AIs. Just like, you know, we've all learned over the millennia how to be good parents or how to manage and mentor our employees. We're going to have to have that same job now for AI. And then finally, I think even if we assume a world where AI does all of those things and we're just helping to do its job better, there will never be a world where we aren't involving customer experience with customers, which are humans and I think that's about us developing the strategies that the AI doesn't have the human experience and knowledge to understand. It's about communicating why we're doing things. It's about reacting to crisis in a way that only humans can do. I think there's still going to be a need for the human to make sure that we keep an eye on all the things that AI does well and make sure it gets better. But we have to kind of always be that last mile of human connection in everything that we do. Because AI is going to not yet be ready to do everything. At least, I think for the next 10, 15, 20 years, who knows?
- [10:09]
B
Yeah, yeah. That general knowledge of all things and the human ability, at least for the time being, to tie all those abstract concepts together and stuff like that. That's. Yeah. AI still has a tough time with that. That's right.
- [10:24]
A
And it's constantly learning, but it's never done. Right. I think that's the big realization a lot of us have right now.
- [10:30]
B
Yeah. So how does this shift, you know, again, back to the dashboards? You know, dashboards make us reliant on a lot of things like lagging indicators and things like that, which, you know, good. But, you know, with real time data, with AI's ability to more rapidly interpret things, you know, how does this shift the way that we look at measurements and even possibly measures of success?
- [10:54]
A
Yeah, no, that is a very, I think, provocative question. And I'll give you a provocative answer. I do think that measurements will, by the nature of the evolution of how people kind of do cx, those measurements are going to change. Right. At the end of the day, when you think about something as fundamental in our CX sort of lexicon as nps. Right. Net Promoter Score. The Net Promoter Score is about a human answering the question how likely you are to recommend something based on your experience with a product or service. In a world where the AI can infer your likelihood to recommend and can infer the drivers of that likelihood being positive or negative, do you really need to ask the question? Right. Should you just go straight to the things you should fix? Right, right. And then look at the actual metrics that are a validation of that is, you know, customer attachment, customer growth, customer revenue, customer lifetime value. You may not need NPS anymore. Right. There's a scenario where you may not be needing to survey your customers. Another, you know, another provocative question to ask is, let's imagine a world where there are AI agents in a company and AI agents that we use as consumers. Right. And If I want to do business with somebody, I don't actually even talk to another human or make a phone call or walk into a store. I tell my agent to do my work for me and it goes to the AI agent for the store and it's a computer to computer interaction. So my actual customer experience is basically telling my CX Alexa to go do work for me and then I have no contact with the company. Should I be surveyed on that based on all the traditional measures of customer experience? Because they didn't exist, we've essentially disintermediated the customer from the company through a bunch of bots there. I think you're going to have to find other ways to figure out how well you're doing.
- [12:45]
B
Yeah, yeah. Or, or even to reinforce the value of brand and product and, and things like that. Because the, the, that whole, to your point, the whole middle part of the experience is I'm, I'm trying to attract a bot, not necessarily a human.
- [13:01]
A
That's right.
- [13:02]
B
Acting on behalf of a human. But yeah, yeah, it's undiscovered territory. Right.
- [13:09]
A
Yeah. And we're going to have to kind of take this one day at a time. Honestly, I don't think we fully know all the answers to these questions, but I think what we can be sure of is the measures will change. They will.
- [13:21]
B
You know that moment when marketing wants a landing page, design mocks it up and engineering says, yeah, we'll get to it. Thousands of businesses from early stage startups to Fortune 500s are choosing to build their websites in Framer with where changes take minutes instead of days. Framer is a website builder that works like your team's favorite design tool. With real time collaboration, a robust CMS with everything you need for great SEO, and advanced analytics that include integrated AB testing, your designers and marketers are empowered to build and maximize your dot com. From day one, changes to your Framer site go live to the web in seconds with one click without help from engineering. Framer is also an enterprise solution, giving brands like Perplexity, Miro and Mixpanel the confidence they need to build their websites in Framer. Learn how you can get more out of your.com from a framer specialist or get started building for free today@framer.com Agile for 30% off a Framer Pro annual plan. That's framer.com Agile for 30 percent off framer.com Agile rules and restrictions may apply, you know, even before some of those bigger and larger scale changes happen. You know, certainly lots to talk about. I mean Obviously, you know, I'm talking with, with you from Medallia. Lots of, lots of companies focused on this, lots of brands, really highly focused on customer experience and admittedly a lot of companies doing well in this regard and improving. And yet, you know, I was just at Forrester's conference in Nashville. They reported like the third year in a row, index is on a decline.
- [14:57]
A
Right.
- [14:57]
B
You know, for so much talk about, and again, a few really good examples, it doesn't really appear to be improving overall. Is AI a factor here? Is it making things better? Are there other factors at play?
- [15:14]
A
Well, I think what you find is that, and again, maybe I'll even go more global than this, that when you're in a period of dynamic change, there's disruption, expectations don't get met. You might be evolving towards a path of a better model or a better customer experience set of outcomes eventually, but change creates misalignment of expectations. And I think it's safe to say that because there's a bunch of changes happening, often driven by AI and by people innovating and trying to try new things, that you're going to see bumps on the kind of the CX road that's happening right now. But that doesn't mean CX doesn't matter, right? Because if you think about, just think about customer experiences, there will always be bumps in the roads. I, you know, I liken being a practitioner to playing a game of whack a mole. If you remember that old arcade game where you have a big hammer and you, you try to hit the moles when they come up and another one keeps cropping up, that's basically customer experience, right? There's always another mole coming up that you need to whack. And our job is not to basically continuously focus, solve all these problems until they're gone, because they're never gone, right. There's always a new product being introduced, there's always a new technology disruptor kind of changing the way we do our business. There's always a new crisis that we didn't plan for that happens we need to respond to. So the key is just continuously focus on improving. Sometimes you'll do better, sometimes you'll do worse, but that doesn't mean that you stop, right?
- [16:39]
B
Yeah, yeah, yeah, definitely. So I want to talk and go back to the kind of the shift from dashboards to AI based reporting and things like that. And so, you know, you've talked about how CX professionals are becoming and probably will become even more like AI trainers than analysts. You know, in order for AI to provide those insights and stuff, they've got to be trained on the right information, you know, garbage in, garbage out, like all of that kind of stuff. What kinds of knowledge and, you know, company knowledge, other things like that do teams need to feed these AI systems to make them most effective?
- [17:18]
A
So I think, yeah, I've been thinking about this a lot as we've been thinking about moving even the medallion capability from being more of an insights platform to being more of an actioning platform. You have to ask yourself, what are the scenarios of inputs and outcomes or requests and actions that you need to take? They fall into a few categories. The first is teaching AI not just how to recognize good and bad, you know, interactions that need to be improved, but teaching it how to recognize how to respond to experiences. Right. What experiences should you resolve right away? What should you escalate? What should you route to a product person or a service person or to a process that's like automated through, like a robotic automation process? So I think we're going to want to think about not just how to classify feedback, but ultimately how to think about which actions need to be taken based on what kinds of feedback come in. And that's going to be a human process improvement training exercise. The second is, and this is something I've learned as I've been working on the board of a company that's in the AI model business, that there's always a desire to make the models more accurate. And accuracy comes into a couple of different categories. One, is the model properly recognizing the thing that you want it to recognize, Right. Versus making a wrong determination? Is it inferring something based on not enough information, which basically is the equivalent of hallucinating. Right. And so sort of make sure that if it's hallucinating because it doesn't have enough knowledge to infer the right thing, you got to give it more data. And then the third is, is it giving an appropriate answer? Right. An appropriate or action. And that's really about defining best practices, brand standards, you know, brand promises. You need to, you need to make sure that the model is aware of those things that are uniquely about how you define your business. Right. So that it conforms to those quality standards. And those are all training exercises that we will have to think about as we're taking what today are, you know, manual processes and using AI to kind of increasingly automate those customer interactions and experiences.
- [19:23]
B
Yeah, yeah. So how does that, how does that shift the way that teams are structured? And, you know, now, now you've got, I mean, I, I think of AI as my intern. I mean, you know, with even running the show, it's like I've got, I've got. So I have a team that helps me as well, of humans, but I also have a ton of automation, you know, and in my consulting work as well, you know, we're always looking at ways to automate and low code, no code, like all that kind of stuff. How should a CX team be thinking about the future of structure and staff and all that?
- [19:56]
A
So, I mean, I'd be lying if I said I have like the answer. I don't really know. I think the answer probably lives in a paradigm similar to how you mentioned using tools to help get things done. When people are building software nowadays, they don't just program, they do what's called vibe coding, right? Which is the concept of using your own brains to kind of build things and start things, but then using the AI to fix things or to do the busy work or to test things. I think we're going to move into a kind of a vibe model for a lot of business processes, including customer experience, right. Where increasingly we're going to ask the AI to either finish the things we start or fix the things we didn't do. Right? And we will have essentially a little assistant right on our shoulder, which is AI that we use. Individual people will do that. I think teams will increasingly kind of be man machine teams where we work together on the stuff that we're good at as humans. But we then use the AI agents to sort of help us either do our jobs better or increasingly take parts of our job. And I think it's going to be more of a. Not so much changing the way teams work, but I think it's going to be more of a cyborg model, right, where we each learn how to use AI to the best of our ability. As we get good at that, will they. Will the models and the team dynamics change? I mean, maybe that's probably beyond my pay grade, I'm not quite sure. But I do know that we all have to embrace AI. I think that that's something we can't afford not to do in our professions at cx.
- [21:26]
B
And along those lines, I mean, you know, I think during COVID I know there was a huge amount of expenditure in just digital transformation and, you know, increasing that. Some of that was by necessity, you know, in person, things just simply weren't a possibility for, for a while. I feel like there's a similar thing here with AI adoption. Right. And I think you've you've said some, some things along these lines as well of you know, early adoption like are to a place where the leaders and the laggards, I mean it's there, there already was kind of that going on on the digital experience side again, you know, the, the post Covid stuff. But are, are we in a similar situation with AI and just the speed to adopt is, is one of the critical indicators of success?
- [22:14]
A
I think so. I mean, you know, I was thinking about analogies in the past about, you know, companies that embrace new technologies and don't. And, and, and actually, even if you go back 10 or 20 years ago.
- [22:24]
B
Before.
- [22:26]
A
First wave of kind of AI capabilities came into certain industries very quickly, you saw the leaders win and the laggards basically struggle or fail. Right. And good examples are Netflix versus Blockbuster. Right. Netflix adopted not just mailing CDs or DVDs and then moving to streaming, but realistically, recommendations engines which were essentially an early form of the AI we all use today, made a huge difference right? At creating kind of more loyalty and more usage. Google versus Yahoo. Right. Google came up with a better way of searching and Yahoo didn't and they won and Yahoo didn't.
- [23:02]
B
Right.
- [23:02]
A
And even in the retail world, companies that are just trying to squeeze margin and kind of optimize the costs, they're not going to succeed if they don't use AI to be accelerants for evolution of their business. I think that's just the truth.
- [23:17]
B
Yeah, and is it? I think you kind of touched on it. It's not just optimizing for efficiency alone. Is it a more holistic approach or what's your advice to maybe those organizations that are doing things and they might be doing things in pockets, but how should they really get started in a meaningful way?
- [23:43]
A
Let me answer it two ways because I think I'll give you some ideas on if you're in a particular role, how do you think about using AI in some of those roles? And then, and more generally from a process perspective, how do you think about kind of starting and getting success? I think in CX today there's sort of three, three or four major roles that I see over and over again in businesses. There's the customer facing roles, there's analyst roles, and then there's, I would call them channel management, right. Digital contact center, et cetera, that are kind of empowered by a certain communication technology or channel. I think it's important to look at those channels or those roles and say how can I inject AI to make the people in those jobs do better or to make the processes which are automated. Think about a digital journey or our contact center, phone tree or IVR voice agent. How do you make that better? There's ways to inject AI to make the process better and then to make the analysis and recommendations for either in the moment improvement better or kind of big picture outer loop improvement of processes and systems so that you make the customer friction go away and the customer that needs to talk to you no longer needs to talk to you or the product that's not working is fixed so that they don't have to return it or complain about it. So use AI to do that analysis at scale. And I think you need to think about AI in both the facing roles, the channel roles and then sort of the, the analyst roles. Right. But more broadly, if I think about how to get started, I know this is kind of change management lessons I've learned over the last 20, 30 years is start small, don't try to boil the ocean in your business. I think it's important to find in any organization coalitions of willing participants that want to try something new. Because in any organization, particularly in big organizations, you're going to have people that want to lean forward and try new things and you're going to have people that just don't want to be bothered. Right. Don't force the people that don't want to be bothered to try something uncomfortable and new. Focus on the folks that want to want to try something new. And I think the other thing is continuously educate and evangelize and teach people in your organization when you have something that works. Right. And then the last thing, and this is a self serving comment I'll make just I preface that, but work with companies and partners that can help you be successful. Companies like frankly Medallia. Right. We, we have a lot of experience. We can impart a lot of best practices because we've been doing this now for several years.
- [26:09]
B
Yeah, yeah. Love it. Well Sid, thanks so much for joining today. All your ideas and insights. One last question for you. I like to ask everybody on the show what do you do to stay agile in your role and how do you find a way to do it consistently?
- [26:23]
A
So I, I've been kind of intrigued by AI for a long time. Even before this current round of technology. I've been reading up on it. I, I, you know, I look for that can educate me on technologies and kind of new innovations. I play with technology all the time. I taught myself how to vibe code. Not well but just because I wanted to know what the fuss was all about. I've got two kids that are actually pretty good techies, probably even more technically proficient than me, so they're teaching me. And then I also just have a network of what I describe as my fellow AI nerds that I catch up with and talk to and they show me things and I show them things. I think it's just important to kind of immerse yourself in stuff you like and care about so you can stay fresh.
- [27:07]
B
Yeah. Love it. Well, again, I'd like to thank Sid Banerjee, Chief Strategy Officer at Medallia, for joining the show. You can learn more about Sid and Medallia by following the links in the show notes. Thanks again for listening to this episode. We're resuming brand new episodes after the new year starts, but thanks for tuning in and making this a great 2025. 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.gregkillstrom.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.
- [28:16]
A
The agile brand.