2959: PwC Explores the Future of AI and Customer-Centric Marketing

[00:00:00] How can companies balance the need for personalization with privacy concerns in today's data-driven world? Well, today I'm thrilled to have Brad Herndon from PwC with me on the podcast to explore this and much more because Brad brings a wealth of experience in customer-centric strategies,

[00:00:21] data quality, and the practical adoption of AI and analytics. So in today's episode, we'll delve into the challenges of ensuring data quality and consistency across customer touchpoints, the shift towards audience-based marketing, and also how AI can help automate tasks to improve efficiency.

[00:00:42] Now today's episode is made possible thanks to my sponsor who helps me cover the expense of producing this daily tech podcast and ensuring that you get 365 interviews at least a year from me. So let's state the obvious here, legacy DRM failed to securely enable external collaboration

[00:01:01] on sensitive files, especially now we're in a hybrid working world. And as a result, organizations face a risk-trust contradiction. Yeah, they must share content with untrusted third parties and yet are also challenged with protecting that data.

[00:01:17] So maybe it's time for a modern DRM solution that solves that dilemma without compromising security or productivity. So if you want to share files without losing control, a company called Kiteworks lets you stream editable, zero-latency video renditions you can enable co-authoring while keeping

[00:01:37] your original files protected in a secure enclave. No file transfers or decryption needed, simply secure collaboration with Kiteworks today where you can say goodbye to deployment headaches, file transfer risks, collaboration barriers and productivity constraints.

[00:01:55] So why not experience the modern way to collaborate on sensitive content but without sacrificing control or security? You can do all that by visiting kiteworks.com to get started. That's kiteworks.com to get started today. But now let's get today's guest on.

[00:02:12] So buckle up and hold on tight as I beam your ears all the way to the US where we will uncover how organizations can navigate all these complex topics to create an effective and most importantly trustworthy customer experience. So a massive warm welcome to the show, Brad.

[00:02:32] Can you tell everyone listening a little about who you are and what you do? Sure. Thanks for having me, Neil. I'm Brad Herndon. I'm a partner in our marketing transformation practice, part of our customer transformation practice.

[00:02:44] And what that means is we think about customers first and foremost, experiences and me in particular using data and analytics to understand our customers changing needs, wants, desires, hopefully in more cases than not to improve the way in which organizations are contacting

[00:03:03] them, engaging with them, offering them the right products and services. So I come from a data and analytics background starting out on the collection side. I actually started out in agency world. And then over time, as I started to do more with data and technology solutions, moved

[00:03:20] into the consulting world, we've been spending the better part of the last 12, 13 years just working hand in hand with organizations trying to understand what's the best way to make use of data for impact in our organization, but increasingly turning that focus to impact for our customers.

[00:03:38] If we do that well, the rest should follow. I'd love to dig a little bit deeper on some of those transformation practices, because right now, predictably all anyone's talking about is AI. That's the big topic. It was last year and it is this year.

[00:03:50] But at the heart of all that though, AI is useless without data. And I think we've all seen in recent years what happens when you put garbage in, obviously you get garbage out.

[00:04:00] So I've got to ask when you're working with clients, how do you ensure that data quality and consistency across so many various customer touch points in an organization? And any strategies that you recommend to employ that would integrate customer feedback into data driven decision making processes?

[00:04:18] Because it's a huge topic and data is at the heart of all that, isn't it? It's at the heart of all of it and it's been the hurdle for as long as we've been using data. I mean, we had this concept of big data many years ago.

[00:04:34] Now it feels like and the thing that bothered me about that topic in particular is big data was talked about as a solution when in fact, it wasn't. It was a problem. We have so much data.

[00:04:47] We're trying to understand what we should be using, how we should be using it. We're trying to stand up the technology to make that possible. That's a challenge. It's only getting to be a more difficult challenge as time goes on and technology evolves.

[00:05:01] And now we've got and certainly there are components of AI that's a very broad topic in itself. I'm sure Neil, as you are studying and interested in having conversations, you probably feel similarly, which is it's a lot of noise to cut through. There's AI as an umbrella.

[00:05:17] There's ML. There's now Gen AI. And what we're actually talking about, I feel like can sometimes get lost. But back to your question, we're still in new ways trying to establish how that data is used. The power of AI and these models and patterns requires it interpreting something.

[00:05:40] And that something means I have to be able to obtain data. Now we've gotten better at taking data in its raw form and transforming that. And AI certainly been an accelerator in that. But for the most part, we still need to get back to basic.

[00:05:55] What data are we talking about? And I think organizations oftentimes have a different idea of what data is useful for them versus what data may be useful for customers. If we think about what customers want, they want to be their needs met, their desires met.

[00:06:13] They want things to be easy enough for them to get what they need and whether that be products or services. And brands often think about data as, let's listen to what my customers are doing, meaning how they're behaving with us, how they're engaging with our,

[00:06:33] whether it's websites, whether it's in-store, whether it's call center. So I can understand the what, but it should really be more about the context around that. If people engage in that way, what does it mean about what they're looking for? And are we meeting those needs?

[00:06:48] And I think we're progressing towards that still, but I think it's still the primary obstacle for these transformations that we're trying to address. When you're advising of strategies and how to make the boat move faster or move these projects along,

[00:07:04] I'm curious, is data silos still a big thing? Are departments still fiercely protective of those data silos? Is that still a thing or are we moving away from them? It's still very much a thing. Whether or not departments are protective of them,

[00:07:18] or it's just that we have so many that it's now hard to reconcile. I might suggest that it's the latter. I think everyone wants to use data. They want the value from it. They're not necessarily protective other than just making sure it's being used responsibly.

[00:07:35] I think the protectiveness comes from, and starting my career as an analyst, I totally appreciate this one. If I'm providing access to the data, do people know how to use it correctly?

[00:07:45] It's really easy to misinterpret data and to have the data tell you what you want it to tell you versus objectively what the insights should be informing. And so the protectiveness, I think, comes from proper use. Less so than it's mine, you can't have it.

[00:08:02] But because there are so many systems and so many different sources in which we can collect and store data now, the volume of silos certainly has grown substantially in the last 10, 15 years. And I'm curious, in your experience, what have been the most significant changes

[00:08:20] that you've seen in customer data management over the past few years? And how are they impacting customer-centric marketing strategies right here, right now, in this AI era that we're entering? The biggest change we've seen is trying to combine these sources.

[00:08:36] No longer looking at just digital data, just service channel data, just sales data, but starting to join these in a single place to the extent possible. There's been what's referred to as Customer 360, Consumer 360.

[00:08:52] And there's certainly been technologies that have been built around the idea of that being the aspiration and the world that I'm closer to, which is on the front office side or the customer experience side. And that includes digital channels, that includes marketing.

[00:09:09] CDPs, Customer Data Platforms, have been one of those key technologies that organizations have turned to help. And that's because you mentioned the data silos, bringing that data together, joining it in a way that is consistent and in the way that customer profiles exist in a database.

[00:09:30] To be really pragmatic about it is still very hard. And so we have turned to technology as an industry to help solve some of that challenge. Now, that landscape is involved considerably. You've got some of these, what we would call point solutions, which are meant to solve this

[00:09:48] and other challenges around segmentation and distribution of the data to outbound channels, like let's say email or paid channels, websites, etc. And then you've got enterprise solutions where these are the large tech players, as we know them,

[00:10:03] who have these as part of their suite of products to better integrate. And now we've got cloud solutions and what's been referred to as composable CDPs as a way to say, well, leverage your cloud infrastructure

[00:10:17] and we can actually take more of a build approach because at the end of the day, these things end up being custom. I would say what to the latter half of your question has changed is also including data that we may not have traditionally.

[00:10:30] Like I mentioned, it's not just about how people have behaved, but it's about the context behind that. Product data, service data. What am I actually offering to them that helps me understand why they're engaging that way? Is it just because of the product or the service?

[00:10:46] Is it because of the price point? Is it because of the ease of interaction? Is it because of the level of care that we provide to them? That context, I think, has long been missing.

[00:10:57] It's, of course, harder to get to, but that's where we have seen solutions like AI help us better understand the context. And that could be through something like affinities. Do people buy a product because of not just at face value, the product itself,

[00:11:10] but the characteristics of the product? How much of the characteristics are important versus the price point or the convenience? We, of course, see ever-changing consumer preferences around some of those things. Quality, price being the longstanding, but also their new entrance, environmental, social responsibility playing a role.

[00:11:32] And so the goal of the data is to also understand what the consumer, the customer's preferences are that are driving towards the decisions, and that would really help us better meet their needs. So with the rise of AI, some business leaders are understandably cautious

[00:11:49] around unintentionally misusing data or even sleepwalking into creepy territory when using AI for things like personalization. So how would you recommend a company prioritize is meeting their customer demands, while most importantly, maintaining and building trust with their customers? It's a fine line.

[00:12:09] We've been chasing this prophecy of personalization for quite a while. And ever since personalization and direct communication with customers, we've been having this debate about what's too personalized and this notion of the customer's eyes, what's creepy versus helpful. And in research, we get conflicting reports back from consumers.

[00:12:31] They actually want to be told more often than not what a company would recommend they're interested in. There's too much choice available out there. No one wants to walk through an endless aisle. They want to be told, hey, I've engaged with you. What do you recommend for me?

[00:12:48] And so we know that preference still exists. I think many of us as customers, consumers ourselves like that preference. But when it starts to get too individualized is when we start to run into issues. One, because there's a chance of error in everything that we do,

[00:13:05] leveraging AI or not, we can be wrong. At the end of the day, these are still human-powered solutions providing the spark and making sure that using our knowledge and expertise and then letting the data and analytics support that. But that means there's always a potential for error.

[00:13:24] And sending someone the wrong message can have severe consequence. And the consequence being the more tailored I am to them, the more that an error has impact. It can really alienate someone. And the consequence of losing the trust of our customers is great for organizations.

[00:13:45] So what we're seeing is a movement towards let's be personalized. Of course, we want to be able to meet people where they are. We want to still help them find the best product, service, whatever it may be for them.

[00:14:00] But we don't want to increase that potential for risk and we don't want to alienate. So how do we find the balance between personalized at an individual level and personalized at an audience level?

[00:14:11] I'll pick on myself and say, there are plenty of people out there in the world that have similar preferences to me. I don't need any brand that I'm engaging with to individualize only to me. I'm OK with there being 100,000s of people who share my preferences and characteristics.

[00:14:34] And therefore, you want to reach all of us in a similar way? That doesn't matter to me. I'm still individualizing the way I see that offering, that message, whatever it may be. And so this one-to-one versus one-to-many debate, I think we're meeting somewhere in the middle, which

[00:14:51] is to reduce risk for brands and to still meet the needs and desires of consumers, which is individualized, but it doesn't have to be at a one-to-one level. It just has to be real to the way that I engage with you.

[00:15:04] And I'm glad you've raised this topic today, because I always try and buzzsmiths and talk about things that people may get confused with. So can you explain the difference there between people-based and audience-based marketing strategies? And what do you think offers more value to organizations and why?

[00:15:22] Yeah, this question is coming up a lot right now for a couple of reasons, some of which I mentioned. But people-based is the notion of we should really be individualizing. We should be sending in the idea of striving towards one-to-one personalization, an organization interacting

[00:15:41] with one individual based on how they've been interacting with them. In some contexts, that makes sense. Certainly for service, someone's calling me or chatting with me. I need to not be generic about the way in which I'm hearing them and responding to them.

[00:15:58] But for many other channels, like I just gave the example of, there are going to be a shared set of characteristic and attributes. And that's audience level. And so while I need to still allow for one-to-one, the way in which I'm servicing people and providing support

[00:16:18] can be audience level. The other driving force that I didn't hit on directly is consent and preference management and privacy overall, which continues to come to the forefront of this industry because the more we individualize, the sense that customers think something may be creepy is a trust factor.

[00:16:36] And we are changing the way in which we can collect and use data. Some of this being driven by tech companies with the deportation of third-party cookies as a primary example. But also increasingly, regulation is catching up and saying, hey, if we're going to provide personalization, fine.

[00:16:56] People need to consent to that being something that they want. And so having consented ways is one way to personalize. But we also know that personalization helps. It helps our customers, and it helps brands. And so even if they haven't consented to one communication directly

[00:17:15] in the way they're engaging now, they've still engaged with us. And how do we use that to still tailor and offer recommended communications? That's audience-based. Because we're moving in that direction, regulation and privacy are dictating it. That doesn't mean that we can give up because we also

[00:17:36] know that it's still what people want. And so audience-based is a way to do that. Segments greater than 50, greater than 100, greater than 1,000, depending on the situation, is an approach to not individualizing, not being too creepy, but still being able to personalize. It also provides economies of scale.

[00:17:55] Many organizations have strived for one-to-one but haven't actually realized it. It's too hard. There's too much data. There are too many channels. To try to do that consistently is really difficult. And is the juice worth the squeeze is the question.

[00:18:09] We look at it from a business standpoint, ROI. And the answer is often that either we don't know or maybe not. And so do we abandon that aspiration altogether? So if we zoom out for a moment, in your opinion, how can organizations effectively transition

[00:18:28] between people-based and audience-based strategies while minimizing disruptions along the way? I suspect that's the next question that you get asked, Dylan. Yeah, I think we're closer to audience-based than if we were the people-based. It's sort of happened naturally is that the trade-offs.

[00:18:46] And then the technology is the other way we get there. Finding out how I should be grouping and automating the groupings. And ultimately, this is their segmentation of customers in a way that's dynamic. If someone changes their behavior from one day

[00:19:07] to the next, they may switch the segment that they fall into. And therefore, I need to automate the fact that they've switched segments and the journey that they're taking so that if they deviate, not sending them irrelevant messaging. If they've opted out of a certain type of communication,

[00:19:24] if they've acknowledged that they don't want to receive certain messages, I have to be able to know and react and respond to that, both from a risk standpoint, but also to stay relevant. So there are technologies, certainly, if we get into the industry,

[00:19:40] both in terms of the notion of clean rooms, right? Which is this idea that I can put data into an environment from multiple sources and it becomes anonymized in a way that becomes usable at an audience level, masked.

[00:19:54] So I no longer know that Brad Herden is Brad Herden, but then I know that these 50, that Brad Herden is one of, which I don't know, but it's 50 people, all have similar characteristics. And there's my audience that I want to go target

[00:20:08] for advertising purposes, or maybe to email for specific products or service offerings. Well, thank you so much for bringing to life this topic today. And before I let you go, is there anything else you'd like to shine a light on today?

[00:20:22] Myths you want to bust or anything that excites you at the moment or anything you're following or some of those frequently asked questions again, anything else you want to raise? I mean, you brought up AI a couple of times. So maybe I'll bite on that one,

[00:20:34] which is that yes, it's a hot topic. Yes, it will continue to be a hot topic. But I think we often need to ground ourselves in the fact that these solutions take time to develop and figure out the proper use of.

[00:20:50] And the last 18 months was, what are we testing? How are we testing spinning up as many applications for AI as we can so that we can show A, that we're responding and B, that we're early adopters. I think the reality is more of a slow progression

[00:21:07] like anything may revolutionize the world in which we operate. Yes, very likely in different shapes and forms. But some of the most compelling ways in which organizations are using AI, Gen AI in particular is to improve their day-to-day tasks that they've been executing, finding ways to automate those,

[00:21:28] whether it's research or briefing or even initial drafts of content or asking a report a question instead of trying to interpret data to get that question answered. There are subtle ways in which it is very real and very practically helping organizations,

[00:21:48] and they may not be the most exciting ones. But I think that's where this all starts is practical use of what may seem like a basic query in a report. And the more we can learn and understand that, I think the more and better progression we'll have

[00:22:05] into adopting solutions versus trying to revolutionize the way in which we work tomorrow. And I think that's a perfect moment to close our show today. But before I let you go, as well as sharing your insights, I'm gonna cheekily ask you for one final gift for the listeners.

[00:22:22] And that is a book that has inspired you or means something to you that we can add to our Amazon wishlist or a song that means something to you that we can add to our Spotify playlist. Guilty pleasures are allowed,

[00:22:33] but what would you like to leave everyone listening with and why? Maybe I'll keep it a little professional but also about me as well, which I think we all struggle. Coming out of the time in which we're locked at home

[00:22:49] and figuring out how to work in a different environment, focus has been really hard. So hard for me, I know hard for a lot of others. I read a book by Cal Newport called Deep Work. Maybe it's already on your list. But trying to change our mindset

[00:23:06] and really understand problems, work through problems while Cal's an academic and so may not apply as directly to our world of constant meetings all day, every day. I tried to apply this notion of it takes us time to think. It takes us time to ideate.

[00:23:26] And if we don't allow for that time, we're not being humans. We're not using our brains. AI might actually end up taking on more responsibility than we want because if we're just task oriented all day every day, we're not using the power of our brain.

[00:23:45] So really the idea of pulling yourself away, finding the time to think, to learn, to grow both professionally and personally was something that really sat with me. And I think it's something we could all use a reminder of now and as we continue to try to

[00:24:02] defend ourselves in this crazy world. So that'd be my book. Awesome. Well, I'll get that added straight to the wishlist. It doesn't ring any bells. It's certainly something that I would like to check out. So I'll add that to the wishlist.

[00:24:15] And for anyone listening, wanting to find out more information or exploring some of the topics we talked about today, is there anywhere in particular on the PWC website you'd point them to? Yeah, I mean, there are some very relevant topics

[00:24:28] that when we covered today, we do a CMO study. We also have our loyalty study, both available on pwc.com. So I would encourage people to take a look there, find real tangible stats around some of the things I mentioned around consumers' desires, preferences.

[00:24:43] What do they actually want from brands versus what are brands presenting to them? I do feel strongly in this idea that first and foremost, we have a responsibility to send to customers. And if we do the right thing and meeting their desires, then brands will be successful.

[00:24:59] So I encourage people to take a look, the loyalty study and then understand how marketers see the world from our CMO study. Well, I'll make sure there's links added to those so people can find it nice and easily. But just a big thank you for joining me today,

[00:25:13] sharing your invaluable insights from PWC. Certainly given me a deeper understanding of these complex topics and their impact on customer-centric strategies. And I'm sure it has to the listeners too. So thank you for your time today. Thank you so much, Neil. Appreciate you having me.

[00:25:30] I for one have gained a deeper understanding of the intricate balance between personalization and privacy today. And also the importance of data quality and the gradual yet transformative impact of AI on business operations. And for me, Brad's insights really highlighted the strategic shifts and practical steps needed

[00:25:52] to navigate these evolving landscapes. But what are your thoughts on balancing personalization with privacy? How do you see AI transforming your industry? Please share your perspectives and continue this critical dialogue by simply emailing me techblogwriteratoutlook.com, LinkedIn, Instagram, X, just at neilchues. But more than anything,

[00:26:14] just thanks for tuning into today's episode. Just thank you for tuning in to today's episode of Tech Talks Daily. But until next time, stay informed, stay engaged and hopefully I'll be speaking into your ears directly bright and early tomorrow. I'll be the guy clogging up your podcast feed

[00:26:33] with another episode right at the top of your list. So thanks for listening today and until next time, don't be a stranger.