Leadership Beyond the Algorithm - Jon Francis

0:00Jon Francis:Hey, welcome to the CMO Whisper Show. I'm your host, Steve Olenski. Part marketing practitioner, part ad agency veteran, part journalist, I was a writer for Forbes for 10 years. I've had so many insightful conversations over the years with business leaders, to athletes, to celebrities, to of course, CMOs. The only difference now is instead of sharing those insights through written form, I'm doing it this way. My guest this week is John Francis, the newly appointed Chief Experience Officer at State Farm Insurance. John's career reads like a master class in how data, digital, and experience converge to drive transformation.

0:41Jon Francis:At State Farm, he's going to be focused on building smarter, more human digital experience that serve both customers and agents. A mission that sits squarely at the intersection of technology and trust. Now, before joining State Farm, John spent three years at General Motors as Chief Data and Analytics Officer, leading enterprise-wide data strategy and innovation. Prior to that, he served as Chief Digital and Analytics Officer at Starbucks, where he and his team built a centralized engine for analytics, machine learning, and research, powering everything from personalized marketing to AI-driven real estate decisions.

1:20Jon Francis:If that's not enough, John's career has also spanned leadership roles at Nike, Amazon, Expedia, and T-Mobile, making him one of the few executives who's helped shape digital strategy across nearly every major industry. I got to meet John at Advertising Week, which is going on about a month from the time we're recording this. And John was either smart or foolish enough to accept this invitation. Welcome to the show. Yeah, no, Steve, thanks for having me. Excited to have the conversation with you today. I am super pumped, my friend. The 30 seconds we had after your session at Adweek alone was like, I can't wait to have this guy on.

1:58Jon Francis:So I want to really get into some really good topics. And I want to start with the lost art of leadership. Now, we talk so much about technology, disruption, and innovation, but not enough about leadership itself. The kind of leadership that inspires, listens, and builds trust. First question is, what do you think has been lost in modern leadership and a follow-up? How do we get it back? Yeah, no, it's a good question. I've thought a lot about this. And some of this is just through my own failures along the way if I'm being vulnerable. But I think, one, we've gone to a point where we equate leadership more with knowledge and differentiated capabilities and skills.

2:43and as I've observed in my career, that's really not enough. It's not enough to just be super deep in a particular topic, and whether it's data, AI, or marketing, usually what it's about is within the organizations that I've been in, is well great, those are a must-have, and those are the requirements, but true leadership comes in the form of, well, you may be smart and be able to build some incredible capability, But then how do you change, drive the change management within an organization to wire those capabilities into the company? And usually that has nothing to do with how smart you are or how deep of experience you have in a particular topic.

3:31It's about how nuanced you are in terms of how you can win hearts and minds and bring people along. And it's not just the team that are reporting up to you, but also how you drive the right amount of influence to drive that change transformation with the senior leaders above you, but also your peers.

3:50Jon Francis:Yeah. And it's, you know, it's it usually at the core of it, it comes down to building those really critical personal relationships with with those individuals to to create advocacy network to help you bring bring those capabilities to life. It's never enough. And unless like if you're working purely in a technology organization, I suppose it may be enough. But as you think about these heritage brands, certainly the ones that I've worked in, there's a lot about, you know, from to in terms of where an organization was, what they've held true and built their brand on, you know, in some cases for over 50 to 100 years.

4:33well that doesn't just transform overnight if you're bringing in a different perspective or different capabilities so for me a lot of leadership has nothing to do with the the technical or the knowledge base it's about how you can bring others along if that makes sense it does

4:51Jon Francis:it does and you used the word influence and I want to kind of follow that up with um like how do you define great leadership in an age where influence often gets mistaken for impact yeah well Well, I would argue that without influence, there is no impact. And so, you know, the impact is realized through the influence of bringing others along, as I mentioned. And, you know, to be able to then execute and implement capabilities and have the right cross-functional alignment to actually bring those capabilities to life. And the impact for me is always, you know, if you set up tests and learn or experimentation to then take whether it's, you know, some form of personalization or you mentioned the real time, you know, the machine learning driven real estate recommendations we were doing at Starbucks.

5:41Well, you set up tests to then prove, you know, how you, you know, once you've gotten past influence and you've brought others along and you're able to test those things in the wild and actually prove through experimentation or other form of measurement that, yeah, they do indeed are driving the impact that you want to see, whether it's about driving cost reduction, creating efficiency or incremental revenue in some mechanism. And then now you've got a path to scale how you can bring those things to life.

6:13Jon Francis:You've been in a lot of places. You've had a lot of leaders yourself that you reported to, right? When you think back on some of those who shaped you, what are some qualities that stand out most? Yeah, I love that question. And yeah, I've seen about every form and shape of leader and the good and the bad and the ugly. And I think for me, you know, maybe it's different for every person. But there are two words for me that are pretty critical when I think about maybe three. One is empowerment. One is accountability. And the other is emotional intelligence. And I think when I think about empowerment, a lot of people talk about empowerment.

6:52Yeah, like I have a leader who empowers me. But it doesn't go far enough if that leader, and I think a lot about this for myself, for my leadership is empowerment is one thing, but if you're not being clear on the accountability and the expectations of what should be delivered and by when and how, then, yeah, it's great to be empowered. but how are you ultimately ensuring that you're delivering against expectations and goals? So I always find that, you know, it's those leaders that both, yes, are going to empower, but also be really clear on accountability and expectations. And then I would also, as I said, just emotional intelligence really goes a long way.

7:34And maybe, you know, the dawn of, you know, AI running everything and algorithms, you know, we could sort of, the emotional intelligence can get lost. But I think when it really comes to strong leaders, it's about meeting your team and people where they're at, understanding where they may be, maybe where they're having some challenges, where they may need some nudges, maybe where they're getting a bit burned out and you need to be leaning in to give them space to kind of energize. So I actually find, you know, that one in particular for me, you know, some of the worst leaders I've had, just to be direct, have been really light on emotional intelligence.

8:15And it has led me into places where, well, you no longer want to work for an organization where you're just, you know, commoditized and being pushed hard. And so I think there's, you know, a real benefit and value to emotional intelligence. And it's a really a powerful way to separate yourself from other leaders because it does feel like that's getting to be more of a lost life these days.

8:39Jon Francis:I'm a big believer in you learn. It's important to learn what to do and also what not to do. Yes, 100%. And I want you to go back in time for me for a second again. But when you first started leading, is there something that you got, let's say, completely wrong? It's not just one thing. I don't think we have enough time to cover that ground. But, no, I think, you know, I think one of the things, you know, I'll think back to my Starbucks experience, you know, when I first started to get into more senior leadership. And, yeah, I think for me it came back to, you know, maybe not meeting people where they're at.

9:21And, you know, and I was so convinced. And again, you know, from an analytic and data science capability, some of the products, analytic products that we're building and solutions, I was so convinced and it was so obvious to me, you know, how to how to bring those to life and how they can help the organization. I'll give an example. We had built a customer lifetime value model, which not only is really powerful for marketing applications, but also how you can use that to inform product decisions, where you want to start to invest, for instance, in digital and starting to understand, you know, hey, for every dollar you put into product X versus product Y, it can lead to a pretty significant yield, differentiated yield in the lifetime value.

10:07And so for me, it was always obvious. I think the mistake that I made, I'll use a dated movie reference here, but sort of a field of dreams that if you build it, they will come. You know, I thought as I built a lot of capabilities, so obvious to me that, well, you know, you put them out there. And then, of course, marketing teams and product teams and others will start to gravitate to leverage because, again, in my mind, it was always so clear. I think the part that I misjudged and, you know, I've really focused on since is, and it comes back to the change management, is really being intentional and mindful about kind of walking a day in someone else's shoes and understanding their day-to-day and their workflow and how they get things done and making sure that as you're building capabilities, you do it in a way where, well, it's not, you know, you're not throwing things over the fence, and that may seem super obvious, but it's more of a partnership and more of having the humility to understand that.

11:03Well, maybe actually the thing they don't need right away is lifetime value, but it's actually something a bit more basic and you could build towards the more advanced capabilities. But I think it does come back to them, you know, really building strong listening capabilities, more empathy and really trying to understand and meet people where they're at. Yeah.

11:23Jon Francis:And I think right along with that is humility, right? And leading, look, we're in the age of algorithms. Let's not kid each other and that's not going anywhere, right? Right. So humility can very easily and humanity, for that matter, can be put on the back burner. So how do you, John Francis, how do you personally balance confidence in data, data driven decisions with the humility to question what the data might be missing in the first place? Yeah, you know, this is a really good question. And, you know, it'd be, you know, it'd be easy to say data and algorithms, Uber all this. But the reality is, you know, I really try to take more of a long game perspective.

12:01And sometimes, you know, some of these brands that, like take Nike or Starbucks as an example, these are incredible brands in the, almost in the absence of data. And the gut-based decisions that were made from the early days, respectively, of each brand that are, you know, 50 plus years old now, that, yeah, based on gut instinct alone, They built something super incredible, multi-billion dollar beloved customer obsessed organizations. And so it's not as simple. I think one is, you know, coming back to the word we used, humility to say for me as a data professional, it would be very easy to come in and say, well, you're not thinking about this right.

12:41This data is telling us something quite obvious versus me coming in and having the humility to honor and respect what these brands have done almost in the absence of data. and then how can you emote that, if you will, just in terms of how you present or represent the data. And honoring the work that's been done that may be more gut-based decisioning, but lensing your work through that in a way that can be really resonant to help those organizations see that, well, this is an individual who truly respects what we've done as a brand, and it's not dismissive just to say this data is going to tell us something completely different.

13:22but rather it's additive. It's an and not an or. And I find personally, just from a communications perspective, that's really critical. And again, I think it comes back to the word that you use to have the humility and the respect to understand, you know, how this organization got to the place it is. And you want to honor and respect that, but you also want to be able to find then a pathway to say, well, how do we then leverage data to augment and put this thing on stairways and really drive to the next 50 years for that organization?

13:56Jon Francis:So how should audience signals, for example, inform creative and message and format an offer without over-personalizing or creeping people out? Yeah, great question. I think, again, it comes back to, you know, it's a bit of head and heart in the sense that, you know, where you can get the greatest efficiencies. And if you took audience signals as an example, well, you know what you don't want to have happen is a human being, a marketer having to over-rev to think through, well, here's a particular campaign. We're trying to sell this drink or this pair of sneakers or this car, and I'm going to have to think through how best to target, like, who should I include in that campaign?

14:39This is a role like no human being can process the dimensionality of data that a computer can do, an algorithm can do, to best inform how you can best target for that campaign. But where I think an algorithm will gap against a human, a marketer who understands sort of the heart of the brand and the DNA of the brand is to then say, how best do we want to communicate with that customer? And what's the creative that we think can be most resonant for the consumer? And yeah, sure, AI can absolutely help allow you to get scale and to create some level of personalization to some degree, but AI will never be able to really get to that DNA and what the core is.

15:30And, you know, the AI can train on that, you know, that insight from a marketer. But really, I think as you think about the creative aspect, that's where I think more of the heart should come in just in terms of how you think about really communicating with those consumers that the algorithms can really help identify and target.

15:50Jon Francis:Since you brought it up, and this might be the longest I've gone in an interview without bringing up those two letters, but you brought up AI. what's your philosophy on ai and marketing like content acceleration media optimization analytics just kind of pull back the curtain for us on your philosophy on ai and marketing yeah look i you know i think you know maybe um those things have always existed back to the database marketing days and you know and before it became part of the common vernacular and you know the headlines building machine learning or targeting algorithms to identify the best customers to target for campaigns or how you use insights from analytics to inform an optimization or how you do automated bidding through more innovative analytic tactics.

16:45Those things have always existed. So I think for me, the thing that feels different now is the scale through which you can now start to get, and the bar has been lowered in terms of how are we able to leverage those capabilities now through AI and generative capabilities that you didn't need to have an army of data scientists anymore. So there's a real level of efficiency that's being created through these capabilities that allows marketers to continue to scale and leverage those capabilities to just get really smart and more efficient in terms of how you squeeze as much incremental value out of those dollars, those media dollars or overall marketing dollars.

17:32And then on the creative side, yeah, I think the thing that, you know, I'll be consistent in that I do think there's still a super important role for, you know, marketers who sort of live and breathe the heart side, the DNA, a brand and a company. But I do think where I found that as someone who's done a lot of personalization in the past, that the long pull was always creative. And how do you make, like, if you really want to truly get to one-to-one marketing, that the creative part of that, that was always the long pull in terms of getting the variance that you needed to really scale, to do personalization, that is truly contextually relevant for every individual.

18:16So, well, yeah, now I do think that, you know, we're in a phase where we now have those capabilities. And, you know, I think it's still, there's some maturation that still needs to happen. I think I've seen plenty of creative that's been generated from AI that, well, why does a customer have six fingers on one hand? I mean, we're, you know, we're not entirely out of that yet, but we will be soon. And so, but all that to say, I think, you know, given the size of device, like most customers are staring at advertising on a pretty small device, your phone. And so I think there's a place and time where you can use a lot of that AI generated content in a way that's appropriate for the channels that people are consuming.

19:02And then maybe, you know, for your linear TV, it's probably an entirely different story. Right. want to have a different level of creative, but that's where you should be deploying, you know, your, your better, higher quality creative that, you know, your agencies or otherwise might be producing.

19:18Jon Francis:I want to just do a slight pivot for a second here. You've been at so many of these huge brands and I want to get your thoughts on partnerships, right. With other brands. And cause I get asked this a lot and cause I deal with, you know, we've talked to so many CMOs and from your, your lens, how do you ensure partnerships at brand equity and performance lift, not just slapping two logos next to each other? Yeah. Well, and again, I think for me, and you may recall this from the panel I was on, but it's always, the proof is always sort of in the pudding in terms of the, you know, what do you, okay, well, how do you measure being very intentional about when you're going on a path to understand a particular partnership?

20:00How do you have and establish the right measurement criteria? What are those KPIs that you'll use to determine whether there's efficacy in that partnership? And then how do you measure it? And it could be, you know, hopefully all boats float up for both brands and there needs to be some mutual benefit. In other cases, it can be that can be executed through brand equity studies or otherwise. In other cases, it might be more driven by more empirical results. And, for instance, when I was at Starbucks, like whether it was, you know, airlines that we were partnering with where we had shared currency on our loyalty points, but being able to then tap in and run campaigns leveraging both customer files where you can actually start to prove that.

20:48yeah, so, you know, one plus one equals three or whatever it is, but that there's benefit for unlocking your customer file for a partner that they are now starting to create more market show for themselves, or they're able to generate more revenue in some meaningful way. So I do think that, you know, I try to be really disciplined around, you know, sort of the proof is in the pudding, just in terms of being really clear on the measurement and how you're going to do the measure to get to that advocacy.

21:17Jon Francis:Pardon my diversion there, because it just popped into my head going, I got to ask John, because you're so, have all this experience. And again, people want to know, how do you measure the efficacy of a brain partnership? So thank you for that. I want to go back to AI and specifically change management, right? And we all know, look, technology moves fast, not exactly breaking news. People don't always keep up, right? Yeah. So my first question is, what are the biggest mistakes you've seen organizations make when trying to implement like technology or AI and then how as a follow-up or even a subset of that is the hype the hyperbole of so much AI stuff now like how do you yeah you know yeah how does one go this is real this is fake literally yeah uh it's such a great question um you know maybe start unpacking the hype part of it first I think one big mistake a lot of organizations are making right now and chasing the hype.

22:15And, you know, I, you hear a lot about companies saying, well, we need an AI strategy. And for me, I, you know, the joke I make a lot is, well, it's like saying we needed a Microsoft Excel strategy. Well, no, no one of course would say that. Like Microsoft Excel is an enabler for you to do work. And guess what? We may not, you know, if the headlines may say otherwise, but AI is really just a very powerful enabler. enabler. So I think for me, when I see companies doing it right versus when they're not doing it right, is you first start with, well, hold on, time out. What are we trying to solve as an organization?

22:53What are the biggest challenges for the companies? Whether how can we drive more incremental revenue? How do we create more efficiencies or mitigate costs in a meaningful way? Start with the business problem. What it is that you're actually trying to get done, and then figure out what are the tools to bring to bear to help solve those problems. And as I said, like AI is an amazing enabler, but I never would want to start first and foremost with like, hey, here's this shiny object. Let's then figure out what we're going to go apply it to. I think you have to be really clear-eyed and clear-minded on what's your corporate strategy, what are the biggest challenges for the organization, and then start there before, or otherwise you have kind of a tail-wagging dog problem.

23:39The other mistake that I see a lot, and I'm sure there's McKinsey or Bain studies on this, is what percentage of companies can't get beyond pilot phase? And you see a large percentage of companies, and I've seen enough studies at this point, they're kind of stuck at that POC. A couple things on that, that maybe the medicine there is one, don't pilot something that you don't see a technical path to get to fully bright and scale. Because then otherwise, frankly, it's just a shiny object. Like anyone can MacGyver and scrap together a small pilot. But if you don't see a path to production from a technical perspective, then you're just kind of wasting everyone's time.

24:27The second part of that, and I do think it comes back to change management, is if, you know, again, a lot of those pilots fail because someone went off in a room and built something and did the surprise and delight and said, hey, look at what the shiny thing that I built. But you didn't bring other people along on the journey in terms of how they could own this capability that you're building. And, you know, most of the organizations I've worked in, having to work across a lot of different functions to get the right alignment. And so I think having more of a collaborative approach to how you build those pilots and also how you think about ultimately the scale, the pilot is going to really help you move beyond just that initial kind of shiny object phase and actually start to really extract value from your AI applications.

25:19Jon Francis:Talking about like, what have you learned? Because people look, we all know the saying that people are always reluctant to change, right? But what have you learned about helping people adapt to change, but not just process it, but really embrace it? Yeah, I mean, it's got to be about co-ownership. And, you know, and for me, the best, when this works the best, it's when if you're starting on a particular initiative, that's going to require some significant change or different technology or it's AI, starting from scratch to almost build a cross-functional pod. And it's not just the people building the capability.

25:59You can't just be having stakeholder throw requirements over the fence. You have to have almost an embedded pod of all the relevant teams working together single-mindedly on this thing that you're trying to drive change and transformation over. because there's something kind of special about it that once you unlock that co-ownership or that co-authorship, then it's less about someone trying to get me to do something different. It's more about, well, we're kind of all in this together and this is our baby and we're going to bring it to life. So it seems like a really simple answer. And I'm sorry for that.