Episode Transcript
Speaker 0 00:00:00 For context, this interview was recorded in June of 2020 amidst the COVID-19 pandemic.
Speaker 1 00:00:11 We all want a business like Netflix or Amazon prime businesses, where once a customer engages with them, it becomes automatic and part of their lifestyle from then on. But how do you build that forever transaction? Robbie Kellman Baxter has been studying subscription and membership models for nearly 20 years. And in this podcast, she uncovers the secrets and strategies of the membership economy. Join us for subscription stories, true tales from the trenches.
Speaker 0 00:00:39 Welcome to the show. It's your host, Robbie Kellman, Baxter sharing subscription stories with you. And today's guest is Daniel McCarthy. Dan is an assistant professor of marketing at Emory university's <inaudible> school of business among other things. He's an expert on valuing companies by focusing on the lifetime value of their customers, a novel approach at the intersection of marketing and finance, his approach, which has won him. Many accolades is known as customer based corporate valuation or CB CV. His approach values companies from the bottoms up by predicting what those companies customers will do in the future.
Speaker 1 00:01:23 Think very thoughtfully about unit economics and don't ignore the growing body of evidence for why it matters for their ultimate valuation. You really can't escape it. So you might as well be upfront about it and use it as a tool to be the heat seeking missile that you use to measure, manage the value of your company. At the time
Speaker 0 00:01:44 Dan's research has been accepted and published in top tier academic journals, as well as nearly every major financial publication from HBR to the financial times to CFO magazine in 2015, he co founded a predictive analytics company Zodiac, which was later acquired by Nike Dan subsequently co-founded beta equity partners to commercialize his work on CBC V customer based corporate valuation. We're going to be talking about why customer lifetime value. It's such an important and misunderstood metric, how to rethink the way companies are valued by the public markets and what all of this means for subscription businesses. Since we're recording in June of 2020, we're going to discuss corporate valuations in times of great market volatility as well. Welcome to the show, Dan,
Speaker 1 00:02:48 It's great to be here with you, Robbie
Speaker 0 00:02:50 Let's let's dive right in. Can you explain customer based corporate valuation for the lay person?
Speaker 1 00:02:57 Yeah. Customer based corporate and its most
Speaker 2 00:03:00 Basic level is an enlightened way of forecasting a company's future revenues. Uh, but driving that revenue forecast off of what the customers will do. So hopefully it's pretty intuitive that, you know, pretty much every major valuation method starts with some sort of a revenue forecast. And the main thing that we would say is, uh, every dollar of revenue has to come from a customer who's making a purchase. So if I can just take these best in class marketing science models for how customers are acquired over time, how long they stay with the firm, the orders they place and the amount that they spend, that's going to give me unique insight into what that future revenue stream is going to look like. That might've been hard if I didn't have all that customer level visibility. And so we're just going to kind of exploit that accounting identity for all that it's worth and bring all that marketing into our financial model.
Speaker 3 00:03:54 Can you talk a little bit about what, what was missing before? What is the value of this that, you know, what you call it, that marketing information that was missing from past methods of valuing business?
Speaker 2 00:04:08 Yeah, so oftentimes, you know, call it 30 years ago, we just didn't have that view of the customer and what the customer is doing over time. And so our valuation method, you know, we could, we would know that customers are going to be placing those orders, but if we can't actually see what anyone's doing, we can't actually incorporate that into our model. And so if we can't do that, we're going to rely on everything else that we do have access to. We know the number of stores that we have. We know basically if I'm extending loans, I know the number of loans that I'm passing out to consumers. Basically, I'm going to take whatever I can get to, to, to put it into my model. But if I don't have that customer level data, I really can't do this approach. I'd say the big thing that's changed over the past, you know, call it 20 years is we suddenly have this really nice view of what the customers are doing, whether it's data coming in through, through digital, you know, that we can now track and tag what customers are doing online, much better than we ever could in physical stores.
Speaker 2 00:05:11 Uh, but even within physical stores, we're doing a much better job of being able to trace back those purchases to a specific customer IDs. And so, yeah, it just opens up a whole new world of possibilities for, for methods like this.
Speaker 3 00:05:23 Yeah. He gets what I remember. I mean, my, my job between years of business school for a brief nanosecond, I worked in investment banking, uh, doing value, doing value. I, I, it was not, it was not the place for me, although I spent many, many days crying in the bathroom. Um, but that's a, that's another story. Uh, but you know, one of, one of the things that we did is we, we tried, we valued businesses. And a lot of times, as you said, we were valuing it based on, you know, well, if last year, this store, the average revenue for this store was X and we're opening three new stores. Then our revenue's going to be four X, right? Cause we have four stores, um, or whatever it is. And so assuming that every store would have similar revenue, even though let's say the customer behavior might be different, the type of customer coming in might be different. Maybe the early customers that come into a store are different than the ones who come in over time or return and become more profitable. And so I think when you, when you talk about the data that we have now, part of it is about the, the accessibility of that data. And I think part of it is about a changing mindset among businesses to think more and understand more about how the customer, each individual customer's behavior impacts that rolled up quarterly revenue number or rolled up annual revenue number.
Speaker 2 00:06:48 No, I think that's exactly right. And it's actually a really good analogy because if you start playing it out right now, uh, you can see how it just is different now that, uh, you might have a company like Walmart or, um, you know, Williams and Sonoma Williams and Sonoma, they do a lot of business in stores, but they get over 50% of their revenues online now. And so, you know, how do we think about that? And typically they think about that in the same way that we think about how an existing store is going to do in future years. You know, that, uh, well, you know, the store, did you attend 10 million in sales last year? Um, seems like the economy is going to be pretty good. So maybe it's going to grow 4% this year, basically some sort of benchmark for same store sales growth, but it's like, okay, so, so how, how are those sales coming about and where are we getting that 4% number from and know typically it's, again, it's some sort of very top down way of thinking about the world. This is to say, um, you can keep thinking in a top down way, but we still know that all of the sales would come from customers. And so, um, you know, let, let's incorporate that into that, that model for, uh, for, for what the revenue is going to be next.
Speaker 3 00:08:01 It's so interesting. The power of the marketing data in the world of finance, um, you know, back to, you know, when I was at, you know, in, in investment banking and I'd come out of, you know, I did marketing strategy for, I was a Booz Allen consultant focused on marketing intensive companies, and I saw the power and understanding the customer, but it wasn't something that was valued in the same way that it is, that it is today. Um, this kind of whole idea of customer centricity and yet understanding, you know, what a financial person might call unit economics can provide so much more insight about the health of a business model. And I want to ask you about blue apron because I know that's a company that you've looked at in great detail. And what you saw that I think some of the public markets missed is such a good example. Think of the power of CBC V. So could you share a little bit about the, you know, blue apron story?
Speaker 2 00:09:02 Yes. This was back, I think it was early June, 2017 and I had just finished defending my dissertation. Obviously my dissertation topic was customer-based corporate valuation. Um, and it was a, it's kind of like you have this hammer and you start looking for nails and, uh, and someone had, had asked the question, you know, over Twitter actually, you know, so what do you think about blue apron? And, um, yeah, I didn't have kids at that time. Again, I just finished the dissertation. I was like, why the heck not, I'm just going to dive in and see, see what's in there. Um, and what was interesting was there was no, no measures of customer churn in the filing, uh, but it did provide some really interesting data about how customers monetize over time. And, uh, you know, basically there was kind of, I was like, you know, uh, we should be able to run a model on this, you know, we should be able to take this methodology and just port it over to, to blue aprons data.
Speaker 2 00:09:57 And so, um, when I did, uh, they were kind of two main conclusions that kind of popped out of that. You know, the first was that their customer retention was not so good. They had something like 70% of their customers churning after six months. And it's hard to build a durable business when, um, you know, basically you kind of have to reinvent yourself every year or every 15 months technically. Uh, so that was kind of conclusion. Number one, uh, conclusion number two was in the run up to the IPO. They started to spend a lot more money on marketing and, and obviously that was spring on revenue growth and customer growth, but not to the same extent that it was before. So, uh, so basically, you know, I had concluded that their customer acquisition costs was rising very rapidly and it had gone from, you know, call it $60 per customer to more like a hundred or $110 per customer.
Speaker 2 00:10:55 And if you just think about unit level profitability and how, you know, how that changes when you've kind of almost double the amount that's spent to acquire customers that can really change the game. And I think what they were trying to do was they were trying to, to hit certain key trophy metrics. It just so happened at the quarter right before the IPO was the very first time they hit a million active customers and there's something really nice about a million. So I think as they were doing that, they were potentially actually destroying value at the end, which, uh, obviously that that's, that's really not good, even though it is creating revenue growth.
Speaker 3 00:11:33 I love the story, of course. And not just because, you know, blue apron of course, is a subscription business. Your, your customer based corporate valuation is not limited to subscriptions, but the relationship with subscription businesses, I find so useful. And of course, blue apron is a subscription business. It's a meal kit, which with very high variable costs because you know, you're not, you're not sending out streaming content or access to software. You're sending out boxes, filled with food that have shipping costs and you know, what I, what I found so interesting about your analysis about blue apron and way I looked at it was they were, you know, as you said, they were spending more and more on acquisition. So basically they were giving people free meals to experience their offering and people were taking that. So they had very high acquisition costs because they were actually giving away food and shipping it to people.
Speaker 3 00:12:32 And you know, the more you give away the longer, the more months you need to be guaranteed that somebody is going to stay, let's call that, you know, if they needed to stay, let's say seven months, but they're not getting profitable until, you know, they're, they're leaving after six months, but it takes seven months to get to profitability. You can see that they're effectively losing money on every new customer. This is not, as you said, a durable model, it's not, it's not a profitable model. It's not about scaling up. It just doesn't make sense if you really have to spend that much to acquire somebody. But if you're looking at it in aggregate, what you see is we put money in for acquisition and we got customers out the other side and the revenue number that, that big revenue number, didn't tell the story of how people were churning out, um, and what, you know, whether or not each individual customer's profitable.
Speaker 3 00:13:27 So I feel like there's a lesson there for the operator, and there's a lesson there for the investor, for the investor. It's, you know, as you point out trophy metrics, aren't always the best, you know, th they're not, they're not the, they're not the most reliable. If you tell somebody what, what the metric is that you care about, they'll do everything they can to hit that metric. Even if, as you say, it destroys the rest of the, of the business model. And I think for the company it's really important, especially in a subscription business to start by making sure that your unit economics work, that you know, that if you bring in a new customer, they're going to stay long enough to justify the cost of acquisition before you, you know what I think of as turn on your marketing loudspeaker and try to bring a lot more, a lot more customers in
Speaker 2 00:14:18 No, I think that's exactly right. Yeah. I think that, uh, for foreign investors, you know, if you knew nothing else about accompany, except that the revenues were growing a hundred percent a year, well, you know, growing a hundred percent a year is better than growing 20% a year, but, but all else is not equal. And, uh, and we need to take into account, you know, all, all revenue is not equal. And, um, and yeah, you mentioned the example of a software as a service firm, typically a businesses like that when they acquire customers, uh, the following year, oftentimes they make more run revenue from that cohort than the previous year. And that's very, very, very far from the truth with, you know, B to C businesses in general, but specifically me, okay. Companies and the, the software's a service firms typically are, are generating, you know, gross margins of 80%, 85% and not, you know, 25%, which, uh, and that makes a world of difference,
Speaker 3 00:15:18 Which, you know, is another really important point. I think for, for subscription entrepreneurs who are, are listening to this discussion, you know, doing a subscription box just of any kind is much harder than a subscription to digital digital goods, you know, digital services or digital content, because you have such high variable costs because you have to, you know, ship the product and you have to buy the product. It's each new widget. You still have to pay for it. And, you know, I spoke last year, I think you've also spoken at, um, one of the big subscription box conferences. And I feel like a lot of them underestimate those, those costs and how difficult it is to generate profit and also how difficult it is to retain a customer for long enough for you to really enjoy the benefits of subscription, which as you pointed out in the world of software, is that, that long tail, that long relationship from that initial marketing spend, right. That they, that you catch them once and you keep them forever.
Speaker 2 00:16:31 Yeah. So I think, you know, you're absolutely right that, that there are big implications for businesses over and above the fact that obviously businesses want to be mindful of how investors are going to think about them, because ultimately to the extent that these businesses need capital, they need to raise money. Um, they're going to need to speak to those investors and so good to, you know, know how the investors are thinking about them and what the report card might look like. But yet even holding that aside, I think just as, as an operator, you can think of things like retention and customer acquisition costs as a kind of measures of product market fit. And, um, if you know that you're not able to retain most of your customers after some reasonably short period of time, there's something about the business model that may require improving.
Speaker 2 00:17:24 Um, so I think about similarly to kind of how you were describing it as well, when we, when we think about acquisition costs and how it makes sense to kind of structure the business, really what a lot of this work will kind of imply is it really matters how much value you're able to get from customers after they've been acquired. And if you're able to get $300 of marginal profit for every acquired you're doing really well. Yeah. That that's, that that's a very, a very good level to be operating from. Cause that means that obviously, uh, in the early days, when you're getting a lot of your customers organically or through word of mouth, your CAC is going to be zero. But, you know, as you grow, sorry, customer acquisition cost. Yeah. As you grow in scale, you're not going to be able to get all your customers coming into organic channels. So word of mouth anymore, and you're going to be rotating into, you know, paid channel. So you're going to pay for things like Facebook ads and Google ads probably, and, and those channels get pricey. Um, and that doesn't mean your business is bad. It just is the cost of doing business at scale. Oftentimes, um,
Speaker 3 00:18:37 The cost of acquisition goes up. Most of the, you know, you want to think that you'll get more efficient at acquisition, but if you're, if you're relying on paid channels, your cost of acquisition will go up. It only goes down if you're getting word of mouth kind of growth as a, as opposed to, um, you know, paid channels.
Speaker 2 00:18:57 Yeah. I'd say the other area where oftentimes, you know, CAC may generally move down our marketplace businesses where there's some sort of network effect that can be exploited. And just the fact that you have a lot more people on the platform, a lot more people in the market will make it so much more appealing to potential prospects.
Speaker 3 00:19:17 Can you give an example of one of those kinds of businesses that saw their CAC go down as a result of growth on the platform and a network effect?
Speaker 2 00:19:26 Uh, so we had done work on a company called Farfetch and for them, uh, their, their CAC head had gone down over time, modestly, which again, just kind of bucks the trend. Uh, another company actually that we'd done work on was, was Lyft. You know, they're kind of classic marketplace now, certainly, you know, with lift the unit economic picture, it looked okay, but not great. And certainly they were, um, IPOing at a really high valuation. So we ended up, you know, kind of being a bit negative on their valuation, but, uh, but it wasn't because their cap was moving up. Uh, their CAC was, you know, very marketplace like that. Um, it came from kind of higher level and then move to a lower level.
Speaker 3 00:20:07 So, so just to summarize with Lyft as they started and people didn't know about them, and there weren't enough drivers out on the road to give people confidence that they should use it. Getting each new rider was very expensive, but over time, as there were more, more Lyft cars available and more of your kind of friends knew that what it was, it became easier for people to find out about it and have faith in it and sign ups of a cost went down.
Speaker 2 00:20:40 That's exactly right. Yeah. With the marketplace, typically in its infancy, oftentimes it's going to try lubricate the market on both sides. It's going to provide incentives to both the buyers and the sellers and the writers and the people doing the rides. And, um, and, and that's going to cause the CAC to be higher. Uh, but you know, once you, once you become a verb, you know, I'm going to, I'm going to Uber now. Yeah. Then, uh, obviously that's going to create the word of mouth that you need to, uh, to not have to spend nearly as much money, which will make the unit economics a lot better.
Speaker 3 00:21:17 So the messages for people listening is that, um, you should really think logically about, you can probably anticipate which way your cost of acquisition is going to go over time, depending on whether you're counting on word of mouth or a network effect. Um, cause you have some kind of a, of a platform or whether you're going to continue to acquire new customers through, you know, more traditional kinds of advertising paid, paid outreach. And, and that will help both you as an operator, uh, anticipate your costs. But also if you're an investor and you're trying to understand, you know, what some prospective investment, what some entrepreneur is telling you about the, you know, when they kind of spit shine their business, these are good questions to ask, to try to understand what do you expect your CAC to be? How confident are you in terms of the lifetime value of each new customer? Who, who signs up? How are you tracking that those are the kinds of questions that might get you better data to make a more informed assessment of the value of the business.
Speaker 2 00:22:27 Yup, that's right. Yeah. If you're, again, going back to the, the subscription box example. Yeah. If you, um, if you're going to pitch me that your business is great and you're basically turning a, your customers are worth a hundred dollars after they've been acquired and you're not spending anything on paid marketing, you know, so your CAC is zero. You will be profitable, but, but I know for sure, as soon as you start trying to scale that, you know, you're probably gonna move from zero to 70 80, and, and that's going to really erode the incremental profitability of your customers. So, you know, CLV yeah. Obviously all of this, it's just lifetime value. Yeah. It's just all basically customer lifetime value, but it's also kind of this, um, this decomposition of customer lifetime value into what I call the customer acquisition costs, which is what you spend to get the people in the door and the amount of value that you get after the, after the required, you know, what I call the post acquisition value. And so being able to kind of see, see where those numbers have been, how they've been evolving, how the, the marketing budget's been evolving and then just knowing kind of given the nature of the business and where it's looking to go, how should move in the future. I think that can give you a very good kind of prior view of, of how that business is going to do.
Speaker 3 00:23:54 It's more complex or more nuanced. I think that a lot of people understand, I wanted to ask you about, you know, we're, we're, this is June of 2020, you know, some of us are on, you know, third or fourth month of, of, you know, restricted movement, uh, sheltering in place. What have you seen in terms of the impact of that on your model and how well your model can incorporate these kinds of disruptions of revenue and dramatic changes in customer behavior?
Speaker 2 00:24:28 Yeah, I think the way that your friend it was right, that, uh, you know, basically it's kind of these, this dramatic disruption that obviously if we were sitting here in January of 2020, you know, our, our model will not predict that, you know, we, we take historical data, we use it to predict the future. I think as we have worked our way through it, um, it's a framework like this that can really allow us to understand how these companies are doing and be able to answer questions like, you know, is the drop in revenues at a particular company because the company is no longer acquiring new customers, but the existing customers are doing okay. Or is it more broad based than that? And you being able to have those diagnostic conclusions about where the weakness is coming from can have definite implications for how that business is going to kind of muddle its way back, you know, as we kind of move our way slowly, but surely, and back towards some sort of a new normal, I would say the, the impact that COVID has had on consumer behavior has been so randomly different for different types of companies where there's been some narrative about, it's just kind of a great acceleration of what was already in place.
Speaker 2 00:25:39 And, you know, obviously the trend towards digital had been in place before and now it's just continued to accelerate, but it seems a little too Pat to me, you know, I think that there's still significant variation where it wasn't even really due to any fault of a particular company that they might've gotten completely host by COVID, you know, even to the point of, well, you know, uh, we sold shirts, so we did a bit better, you know, we, we sold underwear and thus we're not there on the zoom camera, you know? So we didn't, um, when you have, you know, some very hyper synchronized Omni channel strategy, well, all those stores are suck and wind. Whereas the, the company that really wasn't trying and just sold, e-commerce only, and that they're sitting much prettier than now,
Speaker 3 00:26:27 Dose of luck, I think in this one there, you know, things that you can't predict it, I mean, what it does to theme parks and hospitality versus, you know, like you said, e-commerce and how, you know, King Arthur flour is having a field cause everybody's staying home, making bread, you know?
Speaker 2 00:26:46 Yeah. Home improvements been, you know, much more resilient because people are now at home and they say, Oh, darn, I need a better chair for my home office. And, um, was that because they knew it was coming all along. I mean, this is just kind of a random fluke that because people are at home, you know, the mix of products that they need, uh, has, has significantly changed. So see, I think that, um, you know, certainly there's going to be an element where we're just going to need to kind of properly control for what's going to be more temporary. And I think what's going to be more likely to persist into the future. So I think that the, the CBC framework could be a very helpful way of being able to, to see that play out. But I think that there is going to be an element that we'll need another, another handful of minds to really tell what's going to be more transitory and what's going to be a bit more, you know, more permanent.
Speaker 3 00:27:44 Okay. I, um, we're, we're disrupting up and I have some fun questions for you. What advice do you have for, um, entrepreneurs as someone who has started multiple companies?
Speaker 2 00:27:57 Oh, that's a great question. I think knowing what you're good at and knowing where, where your competence ends is extremely important. I know for myself, I thrive when I'm the geek in the back room, I really don't thrive when I'm kind of the, the day to day leader. Who's calling all the shots and that's just, that's not, not an environment that I do well. And, and I don't think I make those decisions particularly well. So yes. So it's not that you wouldn't want to, you wouldn't want to start a business, obviously I've started to, but it can really help inform the composition of the team and the composition of the skill sets of those team members
Speaker 3 00:28:36 That's wise. And then in terms of your advice for academics, who are thinking about, you know, kind of straddling between their research and the commercial side,
Speaker 2 00:28:52 That's a hard one. Uh, typically in general, I'd recommend that academics don't do it. And actually there's a lot of things that are, are good about it, you know, that you get access to great data. Oftentimes you hear these really interesting problems and so it can help spur on what might be the next great academic paper that you write. Uh, but oftentimes they can encourage kind of quick and dirty ways of solving problems. It's like, yeah, you know, this is like the 70% solution that gets done right now versus the hundred percent solution that takes, you know, a full year to do, um, oftentimes to get a paper published in the top tier academic journal, it's more the latter type of paper that's going to end up succeeding. And so, and so having to kind of think the problem solving decision in these very different ways. Um, yeah, I think that can just lead to too bad muscle memory.
Speaker 3 00:29:50 It's like two different, you have two different goals. Okay. So the last little bit, um, I have a speed round for you. Uh, so just answer with the first thing that comes to your mind. What was the first subscription you ever had?
Speaker 2 00:30:02 Oh, that's actually a really good question. I don't know if I have, it might be something like digital news subscription, I'm thinking.
Speaker 3 00:30:12 Okay. Um, your favorite subscription today?
Speaker 2 00:30:15 Geek answer. Crunchy roll. Yep.
Speaker 3 00:30:18 Yeah. Anyway. Yeah, I know that one. Um, the very well run, uh, subscription, your super power,
Speaker 2 00:30:25 My superpower being a random geek and being unabashedly proud of it.
Speaker 3 00:30:32 What are your colleagues hate and love about working with you?
Speaker 2 00:30:36 Uh, what am I calling hate about working with me? Um, I am kind of idiosyncratic, so yeah, I'm a bit of a perfectionist, uh, so that, that can swing both ways when I'm, you know, you're working with other, other human beings. And then in terms of what, what they love about working with me, certainly I'm kind of a zero or 150% sort of a guy. And so when I'm really into something, I really try and carry my weight and you kind of take, take it all the way to the finish line. So yeah, hopefully that, uh, hopefully that benefits everyone who's involved.
Speaker 3 00:31:12 Yeah. That makes for a good colleague. Uh, and then what is the one thing that you want the, you know, the subscription entrepreneurs and practitioners who are listening to this conversation, what is the one thing you'd like them to take away?
Speaker 2 00:31:25 The one thing that I would like them to take away is to think very thoughtfully about unit economics and don't ignore the growing body of evidence for why it matters for their ultimate valuation. Uh, you really can't escape it. So you might as well be upfront about it and use it as a tool to be the heat seeking missile that you use to measure and manage the value of your company over time.
Speaker 3 00:31:53 Unit economics are the underloved and so important, uh, data, uh, for, for most entrepreneurs. A great point. Um, so thank you very much, Dan McCarthy for spending time with me and sharing your wisdom on subscription stories.
Speaker 2 00:32:12 Well thank you for having me and yeah. Hopefully we'll get a chance to pair up again sometime soon. I know you mentioned that the CFO piece, obviously that was one that we'd done together and I had a lot of fun doing that. So looking forward to hopefully the next one.
Speaker 3 00:32:24 Yeah, yeah, me too. I look forward to collaborating again soon. Thanks for listening everyone. This has been subscription stories today. I was talking with Dan McCarthy of Emory university's
Speaker 0 00:32:40 Goizueta business school to hear more success, stories of entrepreneurs, creating their forever transaction in this new and exciting membership economy. Subscribe to my podcast, wherever you download your podcasts. Also, if you like what you're hearing, please give us a rating and review. They mean so much. Thanks for listening. And for your support
Speaker 4 00:33:02 <inaudible>
Speaker 0 00:33:07 To learn more about Dan, go to Daniel men. That's M I N H mccarthy.com. I'm Robbie Kellman.
Speaker 4 00:33:16 <inaudible>.
