Customer Equity: An Interview with Allison Hartsoe, AI Value Creation Consultant and Author

AH: Yes, yes, it does. But we shouldn't limit our curiosity to just marketing.

SS: Well, of course, a good marketer is interested in many different things.

AH: Yes, yes, exactly.

SS: The interesting part about that is we continue to move in the direction of data driven marketing. There's this requirement to actually be more literate in technology, in analytics, and a lot of what would be typically non-marketing subject areas. And hence, I think the industry is at a bit of a crossroads there. Would you agree?

AH: Yes, I think you've really hit the nail on the head there, Stephen. And in a way, we've created these fractional views of our company around marketing, around sales, around finance, around call center, whatever the element is. And when you think about the roots of business, what really drives a business? It comes down to one simplistic thing. People give you money and you have the right to exist. So why take that angle and break it into a thousand different ways of viewing that transaction? In a way, we should be looking at it very simplistically, which is, where do I get my money from and do I have the right to exist tomorrow?

SS: And that's one of the significant issues facing marketing today, is the factionalization of the discipline. I mean, just look at the divide between Brand Marketing and Performance Marketing today, just as one example. And even within performance marketing, you've got multiple disciplines, whether it's social media, digital advertising, etcetera. I think that's holding marketers back, really from making contributions to organizations. Would you agree? I mean, they can't approach the CEO with one voice because they've got a lot of people demanding pieces of the overall budget.

AH: But I don't think they should. These are two different roles. Brand Marketing has what I call the grease in the wheels. You've got to be known. You've got to be able to get in the game. And if people don't know that you exist, then you can't get in the game. And Performance Marketing, I feel, is the other side of the coin. So when we're measuring Performance Marketing, I don't think we should be trying to measure the power of the brand. I think we should be trying to measure the power of the customer coming through and how does that resonate with my business and the economics of my business?

SS: You said earlier, just before we started this conversation, about leaning over your skis on some of these issues that are now, I think, starting to become, come to the fore. But let me go back to, what, was it just mere gravitational force that drew you into this world of customer analytics? Did you see a vacuum there that you could fill? Was it just by happenstance that you sort of tilted your career in the direction you have?

AH: So, you know, what's funny about that is when I began my career, I started as an entrepreneur, and I started working in public relations for a Carnegie Mellon startup. What I learned in that time period was a little bit about the velocity of a company, of a startup, and a lot about Code and HTML and Tracking and Technology, and just the fundamentals of how do we see what's happening before it occurs. Although this wasn't a time for algorithms, this was just a time for data. And so, as I went forward in my career, the second company got founded with a lot of Silicon Valley, Venture Capital Money. And in that realm, we started to really dig into the data streams. And the question on the table at that time was from companies like Bertelsman, who had bought Napster, and they were trying to figure out, how do we push our product across all of these different geographies and know that we are reaching the right people. And so, we were very early on trying to do some of those backbones and really the flow of data. So, unlike other people in the field who have come at this from either a pure marketing background or a pure financial background, I come at it from a data background. And so, in the early days of my companies, I'd be sitting in the boardroom, and I'd be listening to the conversations from the investors, and they would be talking about the ratios, and they'd be talking about how many deals have you closed? And all the metrics that investors really care about. But then the company after that was completely the opposite. And I went down to the bottom of the analytics pile, trying to understand what is coming out of that data flow. And I worked with Gary Angel1 at the time, who is just a fantastic leader in the digital analytics world, and he had some groundbreaking ideas about how do you understand all the anonymous behaviour of the data. So we're using large data sets, trying to find pockets of value. But what we were missing was the connection to the board reel. And I kept thinking, based on my roots, why don't we have a stronger connection to the boardroom? We have all this good data, but the minute an analyst comes running in and talks about what they've found, they're like speaking Greek. They're speaking another language that the boardroom doesn't understand. So I followed my curiosity, and I went on a bit of a quest, trying to make this bridge between the two. Now, early on, I could see that the customer, the grain of the customer, was that bridge. And as we got more and more customer identity, so through the 2000s, 2010s, we got more and more people, more and more data that started to be a good place to focus because you could make that bridge. But not only in the digital space could you make that bridge, but you needed to understand the language of the financial world. So then I spent a little time understanding private equity with a little bit of background from work. And lo and behold, there's private equity talking about the customer, but talking about it in a much different way, talking about it as averages, talking about it as trends, or like, you know, I'm going to basically say that your industry does 6%, so you should do 6% growth. Very, very large estimations. And that's when I realized this is the place for me, because I can make that bridge between the strategy and the tactical. And I understand what it feels like to be the CEO who's getting pitched by the demand of your board and what you need to do. And I think there's a real value in being able to have a constructive conversation with your investors about why you believe certain actions are the right things to take, but you have to do it in the terms they understand. (12.09)

SS: That's a very powerful explanation. It sort of leads us into this idea of customer equity. Obviously, it's become, well, it certainly was the focus of your work. So let's just go back in time a little bit. Even the language of customer equity first emerged mid-nineties and got codified somewhere around the early 2000s. And all of a sudden, there was an academic literature around this, around this concept. But underlying that is this idea of customer lifetime value, which has been in the toolset of direct marketers, mail order marketers, credit card companies forever. It's been a baked-in concept for 50 or 60 years, but customer equity emerged out of that early work. So let me ask you this. Is the concept of customer equity synonymous with customer lifetime value, or is it broader than that? And I say that because the concept of brand equity on the corporate balance sheet is sort of tucked under this heading of goodwill, sort of hidden there, masked, if you will. Where does customer equity belong on that balance sheet, ultimately? And then we're going to get a little bit into the financial part momentarily, but just your overall perspective on customer equity.

AH: So from my perspective, I think about customer equity as the total of the future customer lifetime values, up to 15 years within the base. So I use that as the strategic North Star. I do not include brand equity. I do not include other things that people may think about, because what I care about is the liquidity of the data. How fast can you take action on what you see? And does that point you in the right direction?

SS: So you do equate it with customer lifetime values which is why we're going to get into a fairly deep conversation around that. Let me just raise one other issue. I don't know if you know Neil Bendel2, but I had him on the podcast last year, and he thinks that the term customer equity, he likes it, but he says it's a bit of a misnomer. And the reason he says that is the strict accounting definition of equity is assets minus liabilities is equal to equity. So he makes the case that doesn't really conform to financial reporting. And he says, well, if accountants won't use it, does that kind of limit it to marketing as a segmentation tool and not really as a way of helping the business steer in the right direction. What's your perspective on that? (14.48)

AH: Yeah, I think there's some truth to what he says. And I didn't understand this until I really got into how do you build a financial model to purchase a business? As I was going through that, I was like, wow, there's a whole lot of flexibility in the numbers when we go to calculate goodwill, which is basically a plug on the balance sheet, or when we go to think about how do we form EBITDA, those numbers are not counting standard numbers. But more than that, the question you asked about assets versus liabilities, what is a liability? I could stick 16 things in there. I could stick 6 things in there, depending on my purposes. And that doesn't help me align with the CFO. Moreover, the CFO's office is the root of I would say conservatism. It's the most risk-averse location in the entire company. So if I were to make an analogy out of the use of really digging into the financials underneath customer lifetime value, it would be akin to taking a Tesla and trying to jam it into a Model T. Because the accounting system is Model T technology. It is 1930’s, it is very old. What we do today is faster, stronger, better. And I think it's very important that we lean into how we can take advantage of those numbers without getting wrapped around the axle, on where does it go on the balance sheet and how do we find the right place and the right expenses? And I know that's a contrarian view, because by definition we need to have a sense of the cost. But I don't think you need to go all the way down to exactly which dollar goes to which line item in order to get there.

SS: But it does speak to the continual divide between finance and marketing and their inability to speak the others language or actually understand where they're coming from, which backs into the whole budgeting issue, doesn't it? Like there's this ongoing distrust, if you will, between the two camps.

AH: I think that's true, but I think that's also why I focus on the CEO and the COO, the operations level, because you have to get above that. You know, we can divide this view in - it's like a dividing cake, right? My kid’s 18th birthday, and, you know, they sit there and cut the cake. And the way a kid cuts the cake is, you know, I'll take a quarter of the cake, Mom and Dad, a tiny slice. So unless you are the executive that's saying, okay, now I want to look at my business this way because what I care about is durability of the business, and I care about the ability to command and exit under my terms. That's a slightly different mentality than I have to prove my budget. I have to take action. So the strategy has to fit the CEO's mindset, and then the tactics roll down as opposed to trying to define all the tactics and roll up to the strategy.

SS: How, in your experience, or even your reading of what's going on in the business today, how oriented though are most CEO's toward this? Because I get the impression, given the state of customer experience as an example, that they're not terribly oriented that way. It's the financialization of decision making in corporations, right? They've been infected to some degree. So it makes it a tough argument, doesn't it?

AH: Yeah. So I call that financial engineering. So Friendly's was a great restaurant that I grew up with and totally enjoyed as a kid. Friendly's was a victim of financial engineering. You had two founders who love their customers with all their heart, and they built this business and grew it to a tremendous asset. And then they had financing come in and they did this, the typical strategy of stripping out the restaurant from the brand, and they made them pay rent, they drove up the expenses, they financially engineered the customers away. So, at the end of the day, Friendly’s goes out of business. Now, the world that I want is a world where there are more Friendly’s out there. And we want those CEO's to be armed with the right information that helps them become the next Friendly’s. What they do today is they miss a lot of signals in the data that is already under their fingertips. And the private equity firms don't see it either, because they're looking at the financial engineering. That's how we get shrinkflation, and that's how we get all of these maneuvers that make people unhappy. Now, add a little bit of AI to that, and suddenly we've got the kind of world that I'm not sure I would be happy in. I want the kind of world that's really focused on making my world, my life, better, easier, more convenient, and having the companies and their products be of service to me, not have me be the product. (20.00)

SS: It's so interesting. You're touching on the heart of what's going on in civil society today. Really, it’s like these two land masses colliding into each other. One, the whole ethos of customer first thinking, and the other is, well, bottom line thinking really is what you're describing.

AH: Yeah, but, you know, it's a balance. You can't do one without the other. The company has to have the right economics to stay in business. And so, there's place for financial engineering. There can be waste on the balance sheet, and you can work it out. But there's also a place for the customer and the way we see the love of the customer on the balance sheet, we see the power of the business. And I don't think we should have one without the other.

SS: Yeah, it's interesting with your Friendly's example, because that's exactly what the Red Lobster has gone through, of course. Right?

AH: Spot on. Spot on. Everybody blames it on the shrimp. And there's example after example. Yeah, I call it, there's the hunger beams fast food menu where they're looking for what time of day is hungry prices by ten cents, fifty cents, whatever during that time period, and you add AI to that, pretty soon you've got your soda vending machine scanning your eyes to determine how thirsty you are and then setting your price.

SS: This could be another whole conversation, by the way, we may have to schedule a second podcast just to cover off some of these subjects, because they are certainly pet subjects from my perspective. So I'm going to move us into the more technical aspect of this conversation because you are an expert in the field and you have a very, very clear understanding of it. So I'm just going to mine your brain for the next little while here. So, first of all, I mentioned there's a massive body of literature around customer lifetime thinking. There's lots of books on managing customers as investments out there that have come forward over the years, and all of them talk about CLV - textbooks today, obviously cover the subject in depth as well, but there are also many formulas, and most marketers, their eyes roll to the back of their head anytime they have to look at a very complex formula. You offered a definition earlier, I'm just going to ask you to, if you don't mind, reiterate, what's your preferred definition of customer lifetime value? If you're asked by a CEO who is sort of new to this subject, that's being exposed for the first time, what's your explanation of it to him?

AH: I would make it fairly simple. I say it's the future values of your customers projected out to 15 years. And so, if you'd like to know when your customers are due to buy and when they might not be due to buy, when they're off cadence, when they're on cadence, and you want to be able to make good decisions about how to tactically go after that revenue. If you want to be able to see around corners for when your business is sliding and where it's sliding, you use that future forward projection. I tend to not use the language of customer lifetime value a lot because I think LTV, CLV, it's very muddy in the market. And I cannot tell you how many times I have pulled up, not necessarily an algorithm, but just even a mathematical model, and they are essentially just doing recency frequency modeling. They're just looking backward. Now, I don't know about you, but I wouldn't want to run my business by looking backward. I have to run it looking forward. So that's one thing I always see, and the other thing I always see is, not always, but sometimes see, is these compressed aggregates. Whenever somebody tries to model the entire customer base in an excel sheet - just, no. Unless you have 100 customers. No, you're not doing it right. You cannot model this in a spreadsheet. You have to get the refinement of each person, each individual, and how the discount rate affects them over time. So we take into discount rate, we take into account future projection, we can take into account margin. And I take a big breath on that one because that's where we start to get a little bit of, okay, how much cost are you going to put in there? So I think it makes sense to put a cost in there, so you get some sense of unit economics. But like I said before, I don’t want to get wrapped around the axle on cost. Just get there. You've got so much goodness in the customer base that you cannot see - why spend six months trying to allocate the costs correctly when those six months could be a competitor passing you?

SS: So you've probably answered about three of the questions I was coming up with. I'm going to go back over it because you just said a big mouthful there. But what I'm taking away from what you just said, because I was going to ask the question, is that an historical model that just looks at past purchases and then tries to estimate potential value, or is it a predictive value based on behavioural analysis? I'm interpreting your answer to the question to say, yeah, it's a behavioural analysis of the customer base. (25.06)

AH: It's always predictive. It's always looking at the past signal to generate future strength.

SS: Right. And there's a lot more that meets the eye there. I want to touch on a couple of other things, though, that you did mention. So you've mentioned this a couple of times, about 15 years being the horizon, if you will - explain to me how you get to the 15 years. I mean, I know other companies, so that they plug in a reasonable horizon, say five years. Some others actually try to work backwards to look at cohorts and the actual lifetime, average lifetime of active years of customers. But you say 15 years definitively. What makes you come up with that number?

AH: Because if I'm the CEO of a company and I want to have negotiating advantage with any operating partner of a private equity firm, I want to take every dollar off the table. The point with which we hit infinity. In other words, it's so far out in the future, one more year doesn't make a difference, is roughly about 15 years. Now you can, 15 years is not a tactical number. Fifteen years is the customer equity number. It's our total strategy. So if I want to say, did I make a difference by doing this action? There's two ways to measure it. You can measure it as what happened within the next two years. Did I pull that revenue forward? Or you can look at, how much value did I put in the total customer base? If I'm going to have a conversation about selling the company, then I'm going to make the case that, that 15 years, like a butterfly effect, moved from $500 million to $750 million. And you, Mr. Buyer, are going to buy it here at this value where it's worth $750 million. And in five more years, what I'm seeing selling you today is going to be worth, let's say, one and a half billion. It's a great conversation to have and to arm the CEO with that kind of information, I think is really good in so many different directions.

SS: So I'm going to get simplistic about this, and I apologize for this in advance. But when I think about this, I think about the reliable recurring revenue that you can expect year over year from your customers. There is the unrealized potential, that is, you've got a share of wallet, so there's unrealized dollars there that are being either spread to other competitors or perhaps not being spent at all. And then there's revenue that can come from new products and services that might not even have been imagined at this point in time. So how in a formula, it's predicting future value, do you weigh those possible revenue sources going forward, especially over a 15 year timeline?

AH: I only look at the goodness of the base. If you cannot create a sustainable company on the backs of your existing customers, you don't really have a company. If you're exhausting your entire customer base in five years, then what you have is an acquisition firm. You don't actually have a business. So, the durability of the business is what I care about. And that's why the subtitle of the book is “Data Driven Strategies for A Sustainable Business”. It doesn't mean sustainable like ESG. It means sustainable as in you're in business consistently and that means subsequently that you can control when you take financing, when you don't.

SS: Yeah, I call it Sustainable Recurring Revenue. The revenue that you can depend on. That's the base of your company right there, and anything coming in. But I do want to get into an idea of acquisition because the other thing floating around that I've read out there, certainly as far as CLV goes, is the, you have a separate CLV model for new customers or acquisition customers or prospects versus existing customers. So that's been one line of thought. What's your perspective on that? In other words, taking into account future customers that you don't even have yet.

AH: So I do pick out new customers in the base as a separate group for special treatment, but I don't look at it as future acquisition of phantom customers. I look at it as, how do I feed a look-alike model? So if these are the pockets where revenue, where solid revenue exists, then what are the features you would engineer underneath that, that go with good revenue? And so I guess I'm a little less speculative in terms of, okay, we think we can grow at 6% and let's go get XYZ customers. Or your rate has been Y percent. That's nice, but in a heartbeat, Colin Kaepernick takes a knee and all of a sudden your business changes a little bit. Or these events consistently happen, even though we call them black swans. There's always something that's going to throw your business off. So I stick to the durability of the customer base and then I go fishing for more people like that. (30.02)

SS: So, I want to dive into another area that's always been a bit of a mystery to me and that is the whole concept of retention rates, which are obviously a key part of the calculation. But here's the thing. Retention rates can vary, right? Like a cohort and as a customer ages, doesn't their, it's just like aging period, right? As you get older and keep passing thresholds, your chances of getting older actually improve. So does the same, does the same apply to a customer base where that calculation has to change as a customer passes certain thresholds? Because the chances of them making it to the next year, and year after, year after that, continues to increase. How do you allow for that in CLV modeling? Or is the answer to that that it's actually based on individual behaviour? Like how do you get to the heart of that retention calculation?

AH: When I run the models, I run them again, and again, and again, and again. So the model doesn't just run once a year or once a quarter, it's running all the time. So, that's already picked up in the model as they change over time. So, you bought two years ago, you haven't bought again. There are other factors at play. You know, you can see that flowing through and you can see that changing. The advantage of that is you don't end up like Nike and Lululemon. You can see it coming when all of a sudden things are dropping in one particular area, and they're not coming back. You've got a little red light on that says there's a problem here. That's hard to pick up in the data when you're just looking at product sales or you're looking at retrospective information.

SS: It's interesting because, there's clients whose numbers are dropping, and it's masked because they've raised prices, right? So their revenues went up. You know, hey, we're happy about that. Meanwhile, their core customers are fleeing to the, to the competition.

AH: And that's so important, Stephen, what you've said is it's the quality of the customer base, and that quality component is what we look for, what we seek. That's what I mean by durability.

SS: Yeah. Now, you mentioned cost a few times, and you're certainly clear on this, that you're leery about cost attribution just because it’s, I'll ask you to answer that question, because the question in my mind is, is, oh, they always say about CLV models take into account variable costs. Well, great. But what about all the fixed costs that exist in a business? How do you actually apply that across the base? Do you just take a net margin calculation and come up with that? Or again, do you go back and do that at individual level? But if you're not taking costs into account, then how do, how is it a net margin calculation, etcetera? That seems to me that's at the heart of CLV modeling, isn't it? That it's based on a net margin, or not?

AH: Yes and no. So, technically, if I'm coming from the Wharton School, I'm saying, yes, this is how we calculate CLV, we've got to put the costs underneath it. This is where I come back to that speed and wrapped around the axle point of view. Let's say for a minute that you spend some time figuring out exactly the costs of your customer base, and you put it into the top line, and you know, you put the top line against the cost. And so now you're operating on a margin basis and you're projecting forward as best you can. Will you see the love of your customers in that base as easily as if you give more weight to the top line? Because I find that for most businesses, an excess of revenue solves a lot of problems. So if we're constantly looking at where revenue is coming in and how we can get more creative about ways that we might provide value to bring more revenue in, that drives value creation. So, yes, you can add costs. Yes, I tend to put them against the product, so I tend to say, okay, if I'm going to model, I'm going to say products have a certain cost against them. Yeah, you can put in all kinds of operational costs against it. But why? Yes, you could be very, very precise. But I'm not talking to the CFO. I'm talking to the CEO, the COO. I'm talking to people who want to move quickly in their business and drive value creation. If I let them get wrapped around the axle about all those different costs, at the end of the day, it might not even matter when they go to sell their company that might not even be a factor.

SS: Isn't that marketing’s conundrum generally, though? This whole marketing ROI, MROI question, the fact is that they - well, take social media as an example, eats up what, 20% of budgets today? Point to a social media person who can actually do ROI on social media - there are costs that can't be ever assigned. And yet, back to the CFO question. Marketers, it continues to be challenged on, why do they deserve 12% or 8% or whatever it is these days of the budget. It's an almost impossible question for them to answer. CLV might have the answer, though. If the market can go and say, I've increased the average value of a customer by X over this period of time, which has … this would seem to me that's the way to go for marketers, isn't it? Is to fall back ... (35.21)

AH: That’s easy.

SS: …never mind campaign by campaign by campaign, which is what they're faced with today, or channel by channel by channel. Let's do it by customer.

AH: Yeah, exactly. And that's a top line calculation. So that's just saying, I ran a campaign. These people were expected to buy, and when we ran the campaign, they did indeed buy. But in addition, this other group that wasn't expected to buy also bought. So I'm lifting the quality of the customer base every year by X amount, and that is a much better way to have a marketing conversation. And then the CFO can worry about how much costs they think are going to be attributed to that. But if I were the marketer, I'd be looking very closely at loyalty programs, coupons, discounting, things that can officially erode a customer base. You know, it's kind of an easy lever to pull. Oh, I need more sales. I'll go push out, you know, an offer, a coupon offer, versus really getting to know the customer base and finding what they need and what they want from you. That's a slightly different calculation.

SS: Well, it's short term versus long term thinking, too. You build relationships over time, not by campaigns. Right? It’s…

AH: That's the word, right? Relationships.

SS: …yeah, well, it's, yes, I think it's, it's part of Robert Blattberg, Getz and Thomas' book, “Customer Equity: Building and Managing Relationships As Valuable Assets.” So I want to get into Fred Reichheld's idea of earned growth rate3. He comes up with this calculation and a lot of things we've been talking about today. He talks about in his latest book, and he talks about building in the value of a customer's referral value into the calculation, which, again, is a great idea. And we know word of mouth works, and even more so today, one could argue, yet still also a little fuzzy.

AH: Yeah.

SS: So, when you think of that idea, do you actually think about that when you're building your models as well?

AH: I don't use that calculation, but I like the approach, and I like Reichheld's customer love methodology. Really what he's trying to do is tap into not just the recurring nature, but how, how much can you count on that recurring nature? And I think the referral value makes sense because you're always looking for word of mouth as the most cost effective acquisition technique. So companies that we know have great referral value tend to do incredibly well. Yet when it comes to a data centric point of view, the referral value is seen by cadence. It's seen by, were you already planning to buy, or were you not planning to buy? And then you could, you know, there are other data tricks you could use to try to understand people who are close to each other. So, for example, I ran one analysis for a PE firm, and they were funny. They're like, we can't tell you what the company is, just run the data. So, okay, I ran the data, and I saw this weird signature that I haven't seen before in a data set. What they had was customers, new and repeats, who were spread out across the United States in a very, somewhat even pattern. And you don't usually see that. You usually see wherever the company is based, has a really hard concentration, and there are other places where it's just like, the referral is happening. And usually they're in the big cities. In this case, like Wyoming, that's a very strong referral signature. And I looked at that and I'm like, what are you guys doing? They were running national TV ads. So the national TV was spiking the awareness, the brand awareness part. And then a person who had already bought the product was also telling someone that was right next door to them. And so you could see one node with a bunch of other nodes right around it that were spreading that factor. Now that's more of a visual as opposed to a measurement. Could you create a measurement out of that? Sure, but I think the salient point is, are you getting referral value? If you are, then that's a consistent, good swing, but you want that referral value to come from your good customers, not your, you know, hey, I got a great deal I figured out how to work with anymore. I'm going to get something that I don't ever plan on buying again. That's a more dangerous approach and I've seen both. (40.05)

SS: Well, we won't get into that conversation today. That's that whole NPS versus loyalty commitment score thing that there's trade offs and trends there. So, another question, just to close off our conversation about CLV, because I do want to get into a couple of other areas in terms of operationalizing it within a business. Should CLV be thought of as an absolute, not exact measure, exactly, but close enough measure, or is it more like a, should be thought of as a relative measure? A way to segment customers, whether it's high, medium or low, whatever tiers you want to create. In other words, let's not get too literal about this prediction. It is really just a way to segregate the good customers from the bad customers. Is that too simplistic an approach?

AH: No, I don't think so at all. I think that is exactly how we want to be thinking about it. If we're projecting outward into the future, I don't know what's going to happen tomorrow. Neither do you. It's a probability. Now it's just like gambling. So, it's a guess and it should be treated as a guess. And again, that's why when you try to put it on the balance sheet, you essentially walk into mark to market accounting. Right. Let's assume that there's value there and we benchmark it and we say, here's where we started. Is it going up? Is it going down? What's causing it to go up or down? That's a great way to unpack your business.

SS: You've actually built a predictive analytics model using AI to do exactly what you've been describing today. Without giving away any of the secret sauce on this obviously it's, obviously you've got some magic going on there. But how does that work exactly? What do you need to execute the model? Very rich data set, deep in transaction history, just exactly what goes into it and what comes out of it that helps the businesses that you work for make it work for them.

AH: I always start with the core data. So you need a date, you need a time, you need what was purchased, how much was spent, and you need a customer identifier. So you need four pieces. Well, three pieces of data technically. But if you're going to do anything at all with it, then you need to add some other fundamentals, like what was the address? Or what was the product that was purchased? Now, where it gets more complicated is, and this is the question every company should be thinking about, what are you going to do with that data? Are you going to drive it into marketing? Because that's an easy first place to understand and go after spending costs. Are you going to drive it into your call center and route people differently? Are you going to drive it into your product mix and think about how you make a better recommendation? Are you going to drive it into your salesforce and think about, is that salesperson really performing or not? So there's a lot of different angles to execute and that controls how many other pieces go alongside the core. But there's nothing that says that you can't start with one or two and build, and build, and build. Because again, my approach is iterative. I'm always running and running and running so that it can continually calculate the why behind it. The more data you give it, the better it gets to why. But it's not going to be the holistic answer without understanding the context of the market overall. So there are limitations to what this can do.

SS: But you need a pretty good historical footprint, I would imagine.

AH: I need about three years worth of data. So I need a decent, the company needs to have product market fit, they need to have three years worth of data and they should be at $20 million or higher in terms of revenue. Now, it's interesting to me when we pick out like, who's a good customer for me? Companies that are in the billions, multibillion dollar age, sometimes get as much value in the same model as companies that are running $20 million. So it does depend a little bit on the kind of company and how data centric they are and how customer centric they are.

SS: Well, let's take a bank as an example. I mean, you could have one CLV for a bank, but they have multiple lines of business, each of which has different behavioural patterns which suggest different lifespans across different products, whether it's credit cards or mortgages or you name it.

AH: Yeah, but that's just a dimension. And here you're touching on product centric versus customer centric thinking. So, a company will often come in thinking about those lines of business, right? But I am not a checking account. I am a person, and I have a lot of different needs. So you have to shift that thinking. You have to make product a subset of the customer calculation. (44.57)

SS: But is it conceivable that built into the model would be projections based on those individual lines of business? If you think of them as individual products?

AH: It's a dimension. So when you run the calculation, you look at the overall customer, and then you describe the customer by things that are inside that value. So they might have product A, B, and C, or maybe they have two and a half products, or some set of products. That product is a secondary model.

SS: OK, I'm going to move on from the technical aspects of that into the, into the business aspects of that. And I'm going to ask a pretty broad question, which is, I mean, how do you operationalize the concept of CLV today? And I think you used the term, make it your North Star. Now, I realize that's a big question, because that touches on a whole bunch of other areas. How do you become more customer centric? How do you adopt customer first thinking, et cetera, et cetera? I mean, this is the financial case for doing all of those things. But just from a practical perspective, is this something you operationalize above marketing, or does it start with marketing and spider out from there? How does this work exactly?

AH: So, this is the path I laid out in the book in “The Age of Customer Equity”. And in the early stages of getting used to data and starting to drive by data, a company is not ready to operationalize CLV, even if they can run it, because they don't have enough secondary information about the customer. So, if you run the model, then what? Right? Now I can't take action because I don't have enough information. So, for example, I didn't track my campaigns, or I don't really know how people are traveling through my assets, my digital assets, or how often they're buying. So let's say that I pick up the information about the customer, and I start to understand who my customers are at a listening level. This is an action that happens deep in the company. So my analysts, my data analysts, my business intelligence analysts are more familiar with what this information is. At some point, I need an executive who's in the middle tier of the company to have the desire to do more experiments within at least one, sometimes two divisions. So let's say that your Director of Marketing says, okay, we've got some good data to work with. Let's start to stand on it, let's start to experiment, and they start asking questions of the data. They start asking for more from it. That tends to lead into something like a data lake, where all of a sudden you need to put all that information together, make the customer the North Star, and drive all the dimensions that go with that customer, that value. When they hit that level of the data lake, they start to move upward in terms of who's owning the customer, who's making the decision. So you get a CDO or you get a CAO, you get somebody who has the ear of the executive team, management team, and budgets get bigger, the tools get bigger, the operations get bigger. As they start to move into the final level, which is the leadership level, something really interesting happens. And I haven't seen this across a lot of companies yet. It's mostly around the startups, the startups that have grown really successfully and very fast. There's a sense of liquidity in the data. So, in that middle stage, what usually happens is I jam all my customer data into these different lakes or dams or ways that I store the information. So, maybe I have Salesforce, and then I have Adobe, and I have all these different big systems, that hold pockets of customer data. The challenge with that is in order to be a leader, you have to sense and respond, sense and respond. You've got to go when you see that action happen. And these tools, in addition to being very expensive, often require specialized knowledge to get the information in and out of it. So the liquidity of the data is locked up in that middle stage. When they move into the leadership stage, they're able to unlock that liquidity and start making operational actions. So to me, operationalizing CLV is not something that everyone is ready for, but it is something that companies can do better today than they could before, using AI, using all of these systems like, Zapier is a great example. You can pick up data in one place and zap it to the other place that is operationalizing the information. Today, the tools are trying to do that, but the tools are still fractionalized. So, no tool can be my business. No tool understands my customers or, or will ever see my business the way I do. So the companies have to own that liquidity at the top level. You know, maybe it's sitting on top of a massive database, but one way or another, they have to understand what's going on through the algorithms, stand on top of it through experiments, and then drive action. If they don't drive action, they never move. (50.17)

SS: Let me take it back to marketing for a second. This is a marketing transformation podcast. How should it affect the way marketing does it’s planning? And I speak to this issue of most marketing departments still organized or driven by what product management wants or obviously the desire to sell more product. It's in charge of demand generation or revenue generation for the most part. But the whole concept of CLV underpins the idea of customer management strategy. Those two things have to go hand in hand where customer management strategies should then supersede in many respects product marketing. From a marketing organizational planning perspective, how should it CLV that is be operationalized?

AH: It's taking that tactical component. So let's say that I have all my customers calculated and I can then dimensionalize all the pieces that go with those customers. So from marketing's perspective, that might be what kind of campaigns did they respond to? What kind of data trails did they leave? What is the voice of the customer? How did they respond to their last survey? Did they have an NPS score? What does their social footprint look like? If I add all that information under the North Star of the customer value, then I can see a little bit about what my good customers are doing. I can understand their profile a little bit better. And that means that the dollars go with the actions. So if I'm a marketer, what I want to do is rack and stack different wins. I want to say I'm going after a customer group that looks like $100 million, and I'm going to, my hypothesis is that I can take that $100 million group and move them upward. So I'm going to take my baseline measurement with customer equity. Then I'm going to take my action, campaign, loyalty, whatever it is, and I'm going to measure it again and I'm going to say, did I move that customer base? And using the same methodology to measure them, I end up with that answer and that's the same number I can go back to the CFO and say, we've increased the value of the customer base by X. Now in that conversation, that is not a 15 year number. That's more of a two year, three year number, depending on your purchase cycle. Because you wouldn't want to say, I'm running marketing, I brought $100 million in and now my budget should be $200 million, right? We're not going to have that kind of crazy conversation.

SS: You're not going to see that growth for five years, but trust me, it's coming. But the answer gets to the heart of the question of how many dollars get assigned to acquisition versus retention. It's the old ADR budget split. And companies today, I would argue, still underfund their customer strategy, customer relationship management activities in favor of acquisition, because the emphasis is on growth and more growth.

AH: Yes, yes, I would say that's true. Most companies don't believe that there might be a limit to how many customers you could acquire. So, in an AI agentic4 world, you might not be able to reach me as easily as you can reach me today. So, what if you could only have your good customers and you had to work twice as hard to get the new ones in, and the new ones might not be good customers, you might spend a lot more time thinking about your current customer base. But that's a future conversation.

SS: Well, I don't know if it's so future. It's happening today, right? The whole fragmentation of media is just getting worse and worse. You see the number of dollars now being spent on social media ads just as a life raft, right? For the whole media business. It's interesting times. I want to touch on two more subjects before we close out our conversation. You've been really generous with your, with your thoughts today, and I really quite value it because it's opened up sort of, it's shone a light on an area that, you know, a lot of people don't pay a lot of attention to. They should be. Obviously. It's at the heart of, of customer strategy. This whole idea, though, of tying CLV calculations, customer equity number calculations, back to market capitalization or company valuation. I had Peter Fader on here on the podcast. He's got a whole business wrapped around this idea. Now, what's your perspective on it? On being able to predict, or at least take your customer equity number calculations and map it back to market capitalization just to see how closely they match? What's your perspective on that? (54.59)

AH: Similar to what I said earlier, where if I'm the CEO and I want to have a good conversation about the value of my business, I want to be armed with a number that gives me the strongest ability to have that conversation and say my customer base is worth X. And if I run the calculations in the way that my algorithms run them, then I'm able to have that number, that future number, and I'm also able to have the sense of why that future number is powerful. Then everything else that's added on top of it, new product sales, new customers like product line extensions and new customers is gravy on top of that nugget. So, I like the approach of being able to have a very defensible, algorithmic approach of here's my value. And then the conversation that happens with the private equity firms or the venture capitalists as more of a, here's what I believe: Tell me why you think that's wrong, as opposed to what happens today, which is the firms come in and they say, okay, we're sorting through all your sales data and all these numbers, and we're going to take it apart and rebuild it this way. And they do kind of a bottom up financial calculation. And in that calculation you do have customers, but it's aggregates and they're making average percentages of, you know, you should be doing like this, you should be doing like that. That leaves the CEOs, they're not armed for a good conversation, because it requires them to be a CFO to understand why those numbers are moving around in different ways and how that affects the valuation of their business. So even though this number might not be an approved accounting measure, I think the ability to negotiate from a position of strength is valuable, and what CEOs need to be bringing to the table to command that conversation. It also tells them whether they really need that much money. But I want to go back to one thing you said before about it's a marketing transformation podcast. What marketers can take away from that is, say that you're a marketer and you've run a lot of dollar value CLV-based experiments and you've learned a lot about the customer base. Well, guess who you're positioned to be? You’re positioned to be the CEO or at least move up into operation, sometimes CAO, then CEO. And I've seen this in a couple places where, because the marketers are close to the data and they're close to the customers with a little bit of twist, like a little bit more knowledge of operations, they end up being an excellent candidate for leadership.

SS: Yeah, they should be, and they often aren't. Because marketing isn't considered to be serious people…

AH: Quantitative.

SS: …by most CEO's today…

AH: Which is not true.

SS: …that’s good.

AH: Marketing is very quantitative.

SS: Absolutely. And there are lots of smart CMOs who are close to the market and understand where the business direction is headed. Again, that's a whole other podcast. Final question for you here, because you close out your book saying this, that customer equity is a way to measure the, quote unquote goodness of a company. Can you elaborate on what you mean by that?

AH: So, we talked about customer equity being the durability of the customer base. If you are constantly satisfying your customers and they're happy with you and you're happy with them, see this kind of ongoing durability that's reflected in the customer equity number. What I don't think I talked about in the book is that that's a little bit of a “U” shaped curve. So one way to see that number really grow is to lock up your customers and never let them out. You know, the AOL walled garden or even banks sometimes make it very difficult to change. So I can increase that number through artificial purposes. But at some point, it's too much. Gambling, same thing, right? If I'm draining my 401K to gamble, I've gone too far with that number. So I like to think about customer equity as a “U” shaped curve where, when we don't pay enough attention to it, we've completely missed the boat in terms of we're overspending, we're not optimizing our business. The efficiency is way too low. And then as we come through the bottom of the “U”, we start getting resonance and efficiencies that are really beautiful. And then as you get to the other side of the “U”, you've pushed too far, and that's the time where innovation has to take over and thinking out of the box. So, Domino's Pizza tracker. I want to deliver the pizza not just to my house, but anywhere you are. That's innovative thinking that comes directly from the customer centric analysis.

SS: Well, this is a, it's a great way to close out the podcast. I want to thank you for being a guest here today. Your book is really good because it uses the guests that you had on your podcast as a proof point for a lot of the conversation that we've had today. And so it's a really good read. Have you got a sequel in mind down the road?

AH: I've been thinking about it. I've been thinking about something that's more AI centered and just what I see coming down the path. Where, in a world of agents and AI, how do you stay on top through customer centric thinking? And what's the cost if you don't?

SS: God, that question's on the mind of a lot of people today, so your timing of that couldn't be perfect. Thank you today for this, Allison, it's a pleasure to have the conversation with you. And so it's been fun.

AH: Thank you Stephen. It's been a pleasure.

That concludes my interview with Allison Hartsoe. As we learned, the quality of the customer base is what ultimately determines the sustainability of a business. Yet companies are still clinging to the company valuation formulas of the past without closely analyzing the future net cash flow of customers. Simply modeling the lifetime value of customers using ballpark estimates is not persuasive enough to make the financial case for a greater investment in customer relationships. Companies need to go to the trouble of developing individual level estimates of future customer value using predictive algorithms that can take advantage of the rich array of data that exists today. But to truly operationalize customer equity, companies must then focus their resources more intensely on unlocking the unrealized spending potential of existing customers.

1 - Gary Angel is currently CEO and Founder at Digital Mortar which provides advanced measurement and analytics tools for optimizing physical spaces. Previously, he managed EY's Digital Analytics Center of Excellence.

2 - Neil Bendle is Associate Professor of Marketing at the Terry College of Business, University of Georgia. In addition he is Director of the Marketing Accountability Standards Board (MASB).

3 - Earned growth rate measures the revenue growth generated by returning customers and their referrals.

4 - Agentic AI are AI systems designed to autonomously pursue complex goals and workflows with limited direct supervision.

Stephen Shaw is the Chief Strategy Officer of Kenna, a marketing solutions provider specializing in delivering a more unified customer experience. He is also the host of the Customer First Thinking podcast. Stephen can be reached via e-mail at sshaw@kenna.