A Podcast About Truth, Science, and the Future of Marketing

DD: All right. So, how did you go from Career Builder to PreciseTarget?

RM: Retail is a really interesting problem to me. It's a $4 trillion industry that's underserved, and to put that into perspective, the movie industry is only 42 billion. Auto is only 500 billion, right? Retail is huge, and they've got a big data problem, and I personally don't think retail is going away. I don't think Amazon will control the world. I actually think what's going to happen is, we talk about cars have become computers with wheels. I think tomorrow's retailer is going to be a data platform with stores.

DD: Oh, totally agree with that one. Yeah.

AD: And it's interesting you talked about... you don't think Amazon is going to take over the world. What would you say is the biggest trend impacting the retail industry today?

RM: At the top of my list is velocity. So, what's been happening while we've been not paying attention, is the flywheel spinning faster and faster. So to put it in real terms, retail is traditionally run on 14 week seasons: 14 weeks for spring, 14 weeks for summer, fall and winter. And that is the rate at which they change their assortment of products. Well, fast fashion came along. Zara now is on a two week cycle. Every two weeks, they change their product.

AD: Yeah, that's insane.

RM: Right?

AD: It is so much faster.

RM: Yeah, it is. And now their store visits have gone sky high because it delights people to have new things every time, right? And so, velocity has changed, but that has also put increased emphasis on automation and AI and data, because you just can't peddle that fast with humans alone.

AD: Right. And it's interesting because you mentioned store visits. We hear so much today about this shift from in-person to online shopping and how, really, the pandemic has only accelerated that change. Do you think that's hype? You know, what does the world of retail look like in a post-vaccine world?

RM: Oh, no, it's absolute change. So, what’s been happening is that channel mix from in store to online was moving in about 1% per year. And right before the pandemic, we were about 90% in-store, 10% online. Well, the pandemic was a step change. We saw 10 years of change happen in a year. So we're now an 80, 20?

AD: It's interesting.

RM: It won't go back.

AD: I was going to jump in and say, one of the things that... in particular, is the audience. I think a lot of people were looking to see older people shift their behavior because maybe people... or a certain demographics who hadn't necessarily adopted that online shopping behavior. And that's one of the things that's accelerated, not just the early adopters, but more of a mass population adoption of that online shopping.

RM: Yeah. It is. And the thing that's... where I think many retailers have it wrong is, the consumer doesn't think omnichannel.

DD: Mm-hmm.

RM: They think brand.

DD: Sure.

RM: Right? They have relationship with you. One day, they use their phone. One day, they use the computer. And one day, they're in your store. Right? And that unification is a real opportunity for the retailer, but it also creates technology challenges, right? Because if I go into the Amazon store, which there's one in my neighborhood, I literally, as I'm checking out, I can look on the app and I can see my receipt, right?

DD: Mm-hmm.

RM: Well, I do that at Nordstrom, guess what? Right? It's like, "Oh, that's a different channel," right? But that's not the way the consumer looks at it.

DD: So do you think that with everything, in terms of shifting into e-commerce, how has that affected the retailer's ability to identify new customers? Has it been made it easier? Has it made it harder? You know, Allyson talked about the draw of all ages and walks of life that have adopted this kind of e-commerce, deliver it to my house because I don't want to go be around people, but has it made it harder and more difficult for the new acquisition process?

RM: Yeah, it's a little bit like if they jumped into a sword fight and they have a pen knife.

DD: Okay.

RM: Right? If they're the traditional retailer, right? So the digital IQ of companies is, for retail brands, is to be a key differentiator, and my view is, they're going to have more data scientists and merchandisers in the future, right? And so, the idea that you're going to capture customers because they're going to be browsing the mall on a Saturday and they need something to do, is gone, right?

DD: Right.

RM: And it is a big data game right now. And the ones that can do it are going to win.

DD: So, it's kinda been a double whammy for retailers, specifically ones you were mentioning, the newer retailers are entering into the digital market, but the double whammy in terms of obviously a pandemic driving people inside their homes and not in stores, but also Apple and Google, both taking control over how identifiers can, and if, they're going to be shared across those party lines.

So, how do you see that impacting a level of personalization and engagement, and maybe ability to track and actively use those data scientists that they're building up and hiring? How has that going to impact them in terms of the cookie IDFA and other perishable identifiers?

RM: Yeah. So if we go back to the night fight and many brands have their pen knives out, that it's almost like the boxing ring they're in has started to shift up and down and side to side, right?

DD: Right.

RM: Right while they're trying to have this sword fight with a pen knife, right? And so IDFA, and the elimination of third-party cookies, I wish I could say it's all based on nobility and noble actions. It's not, right?

DD: Right.

RM: And it's based on protecting walled gardens. And so, now there are solutions to the problem, but that's what companies like Neustar are doing. And that's what companies like some of your competitors do, is to help brands bridge that gap. And I today, to prepare for this podcast, I went on to a major department store website they dropped 23 cookies. 23 cookies they dropped on me in that one visit.

Now, I have to tell those marketing people, when those go away, things are going to be different, right? The ability to retarget me and to reach me and do other things, and the thing that I'm really harping on with my clients right now is, as I'm telling them to do the most basic thing, get your email marketing act together. Right? Because that's an opportunity to get an open and drop to refresh your first-party cookie, right? And we've learned that if you personalize that email, make it highly relevant, they'll open it. People don't hate email.

DD: Right.

RM: They hate bad email, right? And so, it's part of the overall strategy. You have opportunities to reach customers, and sometimes you need third party help. Sometimes you can do it yourself, but when you do it yourself, you better do it right.

AD: So, you talked about those personalized experiences and how email is really the first step. And I think that that makes a lot of sense. What exactly do you think the role of first-party data is in customer engagement, and how do you see retailers going out building that data set?

RM: Yeah. So, every retailer with the exception of Amazon faces the same problem, which is sparse customer data, right? The biggest department store chain in this country that sells apparel does two transactions a year. The average Prime customer is buying a hundred items. So, if you say to a data scientist, "I know two things about this person," right? They're going to say, "Well, good luck with that. That's not going to work," right?

AD: Yeah.

RM: And so, first party data is really important, particularly if it includes identity, right? But it's a foundational thing. It's not the end. And I can't enrich you if you can't identify you. Right? And so, we think the companies have done a really good job identifying their customers and cleaning up their first party data graph, because it does a lot of things, including a lot of your personalized experiences, but also enrich that dataset with other data so you can play in the game that Amazon and Target and Walmart are defining.

DD: And so, in terms of the data sources, what would you say are the most critical for a retailer to collect? So we talked about email, right? That's obviously the easiest way to get into their lives, into their email, onto their phone. But what other types of identifiers particularly, what about offline behaviors, other types of digital behaviors, maybe the use of panels, what do you think the plethora of data points a retailer should be looking to invest in? And is there a good mix for a particular type of retailer?

RM: So, demographic data is a good starting point, right? Append that to your identity data. But the problem is, I don't know about you, but the last time I've been... went to a high school reunion, everyone matched my demographic, but we sure didn't look the same.

So Amazon they have a hundred columns of data on every customer and you have three, right? You need help and you need to bring additional data in. Now one of the, I think flaws in typical retail marketing is based on repurchase algorithms. You bought Golan shoes. I'm going to assume you're going to buy them again. We actually show that doesn't work. For most people under 35, they have negative repurchase rate on brands. They're less likely to buy it.

DD: Once they fall into a demographic category yeah, they keep them in that category for too long. Right? Versus [crosstalk].

RM: That's right. And so what we, so you need richer data about the customer. And what I think of offline data and other things, I view them as attributes that are really good for modeling. Right? Is that if I wanted to model my customer base, if I have eight demographic columns, that's good. If I have 500 columns, I can pick more granular segmentations. And knowing that you play tennis or knowing that you ski or knowing that you drive a hybrid car is information that allows me to differentiate you from another customer, right? In a model.

DD: Right. Yeah.

AD: Improve suggestions, for sure. I mean, that's one of the things that I think Amazon does really well, which is give you those suggestions of what else you might buy, as opposed to repurchasing. You're getting a greater interest from the wide assortment that you may offer to those types of customers.

DD: Yeah. And they're also pushing you down further into a rabbit hole of your own making, right? They're allowing you to refine your own audience segment through the suggested products and services that they're offering to you, I guess, just products. So I really want to touch upon protection.

RM: Yeah. Well, I was just going to say, something that fascinates me about Amazon. I couldn't understand why is their personalization so bad, right? Why do you and I the same sneakers? Right? Well, we don't have the same purchase.

DD: Right.

RM: And then we figured it out. They want to be a media company, right? That their business, last year's revenue, 20 billion in sponsored listings. Right. They want to sell the placement. As opposed to personalize the placement. And that just to me is fascinating.

DD: So I mean, obviously having your own data that you're bringing into the mix, whether it's, you're building a model or whether you're bringing other assets into it. Obviously that's proprietary, that's your secret sauce as a retailer. That's what's going to differentiate you in market. So in terms of protection, what have you seen work or not work for retailers to really keep that data close to the vest, keep it inside their own walls, but also make sure there's no data leakage or any sort of potential issues in terms of compliance.

RM: Yeah. With CCPA and pseudonymisation, this is very topical. And now the choice we made is we actually use Neustar to be the trusted keeper of identity. And that is, so if you hack into us, you can't get to anything. It's you can only get to a pseudonymized ID. That's the true ID is held somewhere else. And so that's important. I'll tell you one of the tests I have for every time I see a hacker report. Anytime I see this bank, this retailer, this party was broken into and they stole the passwords, as they run the other way. Right. Because no one's should ever store clear text password. Right. All passwords should be encrypted in a hash. That way, if someone breaks in, they can only get the hash, they can't get the password.

And I think Twitter recently had, or not recently, the last couple of years, had a breaking, people got passwords. It's like, who the hell are you people to store clear text. And so some of that is best practice. And that's why when you log in and you lost your password, they make you create a new one because they can't decode the hash. They make you redo it.

AD: So how about data collaboration in that environment? Because we're seeing more and more engagement where retailers, either retail at a retailer data sharing, or retailers at a brand data sharing. Do you see an opportunity for greater data sharing in the industry despite those limitations that you just mentioned?

RM: And between retailers and let's say a wholesale brand? Like that kind of thing?

AD: Yeah. Or retailer to retailer. I mean, like you said, all of these retailers are in this environment where they're sort of the data rich and there's the data poor even within the retail industry. So how do we facilitate greater data sharing so that retailers can maximize that opportunity?

RM: Yeah. I think there's a big opportunity. One...and this is not a commercial for Neustar, but you're positioned to help with this. Right. So let's say between a wholesale brand and a retailer, there's natural tension. So let's say arbitrarily, you're Calvin Klein and I'm Nordstrom. Well, you don't want to tell me, I don't want to tell you who my Calvin Klein customers are because I'm afraid you’re going to steal them. You're going to conquest them and go direct. Right? But Calvin Klein's saying to Nordstrom, tell you what, we'll give you co-op dollars if you target the best customers at Nordstrom for my products. Well precise targeting knows both, right? We know all the Nordstrom customers, we know all the Calvin Klein customers.

And what a neutral third party can do is say, let us create a dataset in a third party haven that is de-identified. It's only pseudonymised IDs. No one can see the names. We know the names, right? And so there are ways of doing that. The co-ops tried to do this early on in permissioning, say that you can see my data, but only for the, but they were very limiting. In this case I think it would be better is if there was a neutral party that says, let me broker something for you and protect everyone's privacy. But at the same time, give you both what you want. You want to sell more Calvin Klein, you want to sell more product Nordstrom.

DD: Yeah. And that's where we see the future state. Yeah, exactly. I mean, that's definitely, selfishly that's where we sit in the ecosystem. And we see the opportunity between retailers and CPG to bridge that gap and kind of ease those frictions using data collaboration, I think is really the value there. Right? And I think that even as you start to see what the targets and the Amazons and the Walmarts build up their media cycles or the media empires, that's going to become even more valuable for retailers to kind of start knocking on their doors, kind of becoming a part of those new walled gardens, so to speak.

RM: Yeah. One of my big worries is you really need a lot of data competence to do these kinds of things. And the typical brand that we deal with doesn't have those jobs. And they look to us as an outsourcer or Neustar as an outsourcer, but there is help out there. And I think that the new world is the data science world, they have to play.

AD: I love that. I think you've mentioned a lot about data scientists and their role in the industry in the future. And we've been talking a lot about the customer experience and current customers and how do you grow that lifetime value. But one of the things that I'm curious about, and one of the challenges many retailers have is really expanding their customer base. So how do retailers reach new customers today?

RM: We looked at 5 billion transactions and we tried to figure out why people make decisions. And it really scared us when we saw the negative repurchase rates on brands, right? We said, wow, we thought there were Nike people. And we thought there were Under Armour people. And there really don't seem to be. But then we took a step back and looked at it. So what are we seeing? And we said, wow, people make purchases are remarkably similar to their previous purchase. So as an example, the jeans in your closets are not all the same brands, but they're remarkably similar. Same style, same size, same price. You have white lines, right? For jeans.

So we said, when we figured that, right? And then we did that. So we built this taste graph of the entire universe. And then we said, what if we targeted people by taste. Conversion went up 50%, five, zero percent. Versus a brand repurchase. So we learned is, people, their purchases are driven by their personal taste. If I hold up five sweaters and said, what do you think? You're really answering that one's my taste. Right?

AD: I was just going to say, I was going to say that reminds me of the different platforms that have come into the retail space, from a digital first, digitally native platforms like Stitch Fix and Rent the Runway. I mean, they're really leaning on that data, that type of data in order to make those recommendations, aren't they?

RM: Yeah, they are. And I think they have a lot of human in what they're doing. And for the Nordstroms, Macy's and big companies like that, they're going to have to do with automation. There's too many customers. But the good news is there is good data, right? We have 30 attributes on every skew in America right now. Right? And so we know the feature set of products and we can fit them to people's tastes. And so automation is here.

DD: You think it will continue to be here? So that's a question I have in terms of acquisition and the tools in which the retailers are utilizing acquisition today in the supply chain and with all their MadTECH partners. Do you find that that process, that supply chain process, that acquisition process will drastically change? Will automation increase, decrease, or will it just be something different we don't really know what to guess yet?

RM: Well, so in every podcast, it might be a tradition of mine to offend my host's best customer. So I'll tell you that the, I think the most broken part of the ecosystem in retail is in the agency, right? Because the incentives are misaligned. So your shoe brand, you want to sell more shoes, you go to your agency, what do they want to do?

Sell more shoes. You go to your agency. What do they want to do? Sell more media, because media is how they make money on commissions. In fact, a good ad campaign might make you buy the less media. And so the incentives are misaligned. And we're really encouraging our customers is get involved in acquisition, measure it. Really be focused on targeting and really be focused on why you're targeting people and for what reasons, because the agencies aren't. They're not doing data science. They're selling media.

AD: Yeah. It's interesting. One of the other things that comes up in that kind of environment when you are more actively involved is cost. Do you have any suggestions for retailers about how they can manage those costs for customer acquisition?

RM: Yeah. Well, we're the opposite of AB. We always say start small, right? AB tests the heck out of it. Let's get to the right target set and then scale it. Because the, the most depressing thing is wasted ad dollars, is shooting money. You might as well just be throwing it out the back of your car as you are driving on the road. And that's what bad targeting does. If you target the wrong people, you're wasting money. And I'd rather see a much more narrow model to start. Really, let's prove it. And then once we get that, then scale that. And measurement is really key and it's not natural to the non-nerds of the world. Right? I was with an agency recently and I asked them how they picked the data. And they said, "Well, I pick the one that looks the best in the trade desk data store." And I say, "Well, do you measure it?" And this young gentleman said, "We do not have time for that. We do not have the..." I said, "Well, you guys are crazy. This is crazy land."

AD: Right, right. It's interesting. You said measurement and you've been talking about AB testing. What kind of KPIs do you encourage the retailers to look at? What does success look like for them?

RM: So cost per click and click through rate and conversion rates are all key. We really look at, are you acquiring the customers who fit your brand. And here's why that's important. Someone who buys an on sale product one time and never comes back is a really expensive way to do business. If you acquire a customer who loves your brand and can love your brand, and you can create high lifetime value, you might be willing to pay more for that customer. And what we do is we have a model for our brand, a data model. And when someone's acquired, we hold them up to it and say, "Where does this person score in this brand? Right? And is this someone you should invest in?" And that is, in data science, we call it expert adaptive. What you really want to do is you want to continually check yourself, check your agency, check it, and tune your acquisition based on the results you're seeing against superlative set of the customer you want.

AD: Yeah. We talk a lot at Neustar about incrementality and how do you know for sure that you're driving incrementality with your ad spend? So it's a really similar concept in that it's important for marketers just to make sure that they understand what is really driving success? And how do you know that what you're doing is incrementally growing the brand, as opposed to more of the same? And it's interesting to take that consumer lens on it, and to really say, "How does this fit with what you've already seen as successful for the brand?" If you think about this past year in particular, what do you think some lessons are that retailers have learned, in terms of how they can improve their relationship with the consumers or shoppers? And what do you think has actually hurt those relationships?

RM: Well, see, I think that the inevitable is here. We knew the customers were going to go more online and it's happening. So it's changed fashion. Right? I call it waist-up fashion. Right? I am wearing long pants, not always, in doing my job. But I think that what I really encourage retailers to do is collapse the Omni channel. See, what happened is these grew up in retailers. There were the in-store people, there were the online people, and there were the mobile people. And they treated them like three channels. That's not the way the consumer looks at the world. The consumer views you as one thing. And they just happened to be wherever I am, I'm going to shop with you. Right? And so the relationship with that customer is really important.

Home Depot is a customer that has integrated in store with other mobile app. So as you check out, the receipt goes right in the app. You know the coolest thing about that is for them? I'm using my app in their store. They have my identity now. They know I'm there. And Sephora's doing something really awesome. I don't know if you shop at Sephora, but here's what they did. We used to talk about consumers are going to showroom and on their mobile phone, they're actually shopping somewhere else. Here's what Sephora did. You can try on the makeup in the store with their app. They have your image in the app. It's a virtual you. It's like a VR thing. And you can use a QR code on some makeup and it puts it on your face. Right?

Well, now I'm using the app. You have my identity. I'm in your store. You know I'm in the store. You know what I'm looking for. You're seeing my shopping behaviors. You also know if it looks good on me. So they're creating this environment that the user wants to use the app in the store, and is all synergistic. And I hope the retailers see that as the future is integrate digital, not use it as an alternative.

AD: Right. It's really an improved customer experience. And it's interesting. It kind of reminds me of the advertising world too, because we have similar things when we talk to agencies. Within the brand, someone's managing TV, someone's managing digital, someone else's in charge of social. But really, on the consumer side, you see the brand in all of those places. It's very similar from a retail perspective. It's just about having a unified customer experience which is connected. Because the consumer has no idea that there's three different people managing that they want to know, "This is how I interact with your brand. These are the messages I see across these channels, and across these devices or platforms. And I want to make sure that I have a similar, positive experience across all of them."

RM: That's exactly right. And yeah, you're one entity to the consumer and you should act like it. Right?

DD: And so we talked about the interconnectivity, it seems, of the experience, both digitally, as well as physical. And so I want to kind of come back to the topic of identity. And so I think identity has taken on a lot of definitions over the past few decades, specifically for retailers who I seem to be always on the cutting edge of being able to gain access to more information about consumers. Because I feel like retailers are always going to have greater access to those bits and bites, because they're more built on that relationship between their brand and that consumer, as you said, over time. So what role is identity really playing, in terms of the future state of marketing, specifically for retailers, versus maybe someone in the QSR or CPG space? Is there something unique to retailers and the future state of our industry that identity will play a key role in kind of amplifying their business?

RM: Yeah. It's table stakes. [inaudible 00:32:46] identified the customer. So I can't name this brand. It's a brand you might be wearing. It's one of the biggest ones out there. That I walked into this customer, and they can identify 24% of their customers. I said, "Wow. You don't know who your customer is. How scary is that?" Right. And so-

AD: How do you build a product for them? How do you know what to offer?

RM: I know. And so here's the way I look at it, is if you are a brand and you're going to play in this sword fight, right, against Amazon and Walmart and Target and all the other big guys, you're going to need additional data. You don't have enough, period, right, to play in a data science driven world. I cannot enrich you if you can't identify that customer. If you have some, I can't give you data. And that means you're going to fail. You need data. So it's table stakes. I can't personalize an experience for you if I can't identify you. And I might not be able to reward you. And so I just think that it is table stakes. The brands that are under... Most are under invested here, and it's really a shame.

DD: I don't mean to laugh at that, but it is a shame in terms of... Because again, it's about a lasting relationship, right? It's about reputation. And I feel like when a brand is in the news for doing something incorrectly with data, that's a huge impact to their long term liability of that brand. And so when we think about the relationship between the consumer and the retailer, it's probably the strongest out of any other vertical that may exist other than maybe car, the auto industry. How do you see privacy legislation playing into that? We talked about Apple and Google, but CCPA, CPRA, the recent announcement in Virginia with the passing of their bill. Do you see that, the retailers that you speak with, are they taking this seriously? Are they putting things in place to make sure that compliance is adhered to? And do you think this is going to evolve the relationship between the consumer and the retailer over time?

RM: Maybe, probably, we'll see. See, I think the ad tech industry has created this view through the wrong side of the prism. See, I don't view CCPA as a privacy law. I view it as a control law. It's putting the consumer in control and saying, "You are in control of your identity and your data. And you at any time can demand what I know about you." And so I think it's one of these things were people love Spotify. They love recommendations for music. It's all based on their data, right, and aggregating with everybody else's, and Netflix is doing the same thing. People don't want to give those personalized experiences up because there's a payoff. What turns people off is when they get re-targeted, like they're being stalked. All over the web, they look at a product that they feel like that this product is stalking them. And so consumers I think are, they don't want to be exploited. They want to be empowered. They want to be empowered to make good decisions and do good things. That's the thing that, if I were a brand, I'd really want to say, have a rights agreement, "Here's your rights that you have, and I'm not going to exploit you. In fact, I'm going to put you in control of it. I'm not going to bury it in some terms of service that's three links deep on the website."

AD: I agree. I think that's where the industry is shifting to, with a lot of the recent changes in the recent updates, and with Apple's announcement. I think there will be a lot more push to have consumer awareness of what they're really buying into. But to your point, there are a lot of positives that come out of it as well, such as the personalized experiences. Especially if a consumer chooses to build a trustful relationship with that retailer or that brand, they expect the retailer or the brand to hold up their end of the bargain in terms of making sure that they keep their data safe and they don't exploit that data. But at the same time, with that comes a lot of benefits. I think we in the industry need to make sure we promote that side of it.

RM: Yeah. We did a project for a major sporting goods retailer recently that was really interesting, that they were sending a purchase confirmation transaction. Got your order, it's a confirmation. They wanted to put offers in there, in that email. And so we personalized those offers for the consumer. The open rate went up three X, and the conversion rate doubled. Now, all they did was service the customer. And it drives me insane when the biggest retailers in the world send me the same email they send you. That [crosstalk].

AD: Frequency doesn't help, either. They send it so many times.

DD: I know.

RM: Yeah. That's push the spam button. But I don't push spam on Spotify. There's a quid pro quo, right? I'm getting something out of this. I think if the retailers service the customer with personalization, service them better, they're not going to push people pushing the spam button.

AD: Right.

DD: It's almost like the [responsibleness 00:39:10] of a retailer is becoming a competitive advantage. Right? It's just that easy for a brand to use your data responsibly, and be transparent about how that relationship is being constructed and maintained, versus a retailer who may just not choose to be as transparent. That's just an easy call for, like you said, for you as a consumer, just to say, "Delete, unsubscribe, you've lost my business", just because of the basics of not really being a responsible data handler.

RM: That's right. Well, think about travel. There's a time when we had travel agents, and I remember-

DD: There was a time when we traveled.

RM: Yeah, yeah, yeah. When we did travel. People were really aghast travel agents were going to go away, and we're going to use Expedia. I actually view the intimacy with Expedia far higher than I ever had with a travel agent. They have all my information. They know where I like to stay in New York. They always have those places. They have all my frequent flyer numbers. They have my credit cards. I think there's a certain level of intimacy with this relationship. Do they have all my PII? Absolutely. They know everything about me, about my travels and things like that. But I'm willing to do that because there's a benefit. Is it helps my travel.

AD: Easier.

RM: Easier.

AD: It's more convenient.

RM: More convenient. I'd never want to go back to a travel agent. That's what the retailer has to get to. Is that if the consumer's getting a benefit they're willing to be less private. If you're going to exploit them and re-target them and stalk them.

DD: Right.

AD: What's next? What do you think is the next big thing for the retail world, and the role that Precise Target can play in that space?

RM: I hope that your podcast is a good place to break news, because I'll just tell it all to you now and you [crosstalk]-

DD: Of course.

RM: Yeah no. Here's the thing. Amazon will not win this whole game, right, contrary to popular belief. The brands are actually pulling out of it. It's because you can't make money there. You lose control of your customer. There's a lot of creative creativity left in the world to create fashion and brands and experiences. There will be other places to buy things. But see, I look at it like this. In the last decade we saw advertising transform to something we now call programmatic. There used to be creative people in advertising. Now it's all algorithms and computers.

Advertising online is a big programmatic thing. Retail is going to go programmatic with two week assortment change velocity, and the amount of data that's available, and with artificial intelligence. As an example, that artificial intelligence might not be a robot walking through the store, telling you what to buy, but it probably would do a pretty good job telling that retailer what to assort in their assortment of products. Why is that important? Well in retail, the money's made on the buying, not the selling. If they assort the right products, that's how you win. Data can help there. Tomorrow's retailer is going to be a data platform. It's going to be really exciting. It's going to be exciting for people like us that are in the middle of this, and that can help them. I think that we're going to be living in a higher velocity world of more personalized experiences, that all retailers won't survive but new ones will. The new ones that will, will be data savvy, high digital IQ kinds of companies, and it's going to be good for the consumer.

AD: Mm-hmm, yeah. It sounds like a great place to shop.

RM: Yeah. Well, of course your robot will be doing the shopping for you.

DD: That's fine. I think everyone's going to get used to not interacting with sales associates and people. I think that will be better for worse for some people. The brands who lean into the nature of not wanting to move out of your own home, I think will benefit a vast number of people who come out of this COVID state in a different mindset.

RM: Yeah.

AD: To your point earlier, convenience. I think consumers are getting used to that convenience, and I think those retailers who have those data assets are going to be able to enable that.

RM: I sometimes wonder if you were ... I live in an old neighborhood, and see, if you were designing this neighborhood today would you put rails to deliver a cart every day, right, to deliver the stuff? The UPS truck-

AD: Drones.

RM: ... go up and down my street every day, right, stops at every house.

AD: Exactly. Yeah. Well, Rob, thank you so much for your time. It was great to chat with you. It was fascinating to hear your perspective on the future of retail. Thank you again for your time.

RM: Thank you. It was enjoyable.