[MUSIC] So now we're going to continue our discussion of how the real world and the virtual world interact with each other in online, offline competition. So if you look at the slide there are five companies that I'm going to speak about. And this is actually very exciting for me because these are companies I've had the pleasure of working with in terms of analyzing their data and also doing some research trying to also help out the management team a little bit. First one's a company called netgrocer.com that was really one of the very, very early companies in the Internet, selling non-perishable groceries starting at about May 1997. And I'm going to show you how that business grew and evolved and try and give you a sense of how it is that e-business companies, e-commerce companies, spread themselves over time and over space in particular markets. The second company I'm going to talk about is one of my favorite, of course, is part of Quincy family. That's diapers.com which begin life in 2005 at 1800diapers.com. In fact, you may see a little bit behind me. I'm actually still here in the Quincy warehouse again, backgrounded by 1,200,000 square feet of warehouse space, robots and so on shipping diapers and all kinds of stuff all the over United States. The third company we're going to talk about, warbyparker.com, this company was founded a little bit later. This company was founded in 2010, in February, by four students of the Wharton School. They wanted to do something very, very interesting. They wanted to take a product that most of us were buying offline, if you wear the product, which is glasses, that's what Warby Parker sells. You would go in to an opticians, get your eyes tested, maybe try on a bunch of different frames and buy glasses in that environment. Now what the forefounders notices is that at that point in time, less than 2% of that product category in the United States was sold online. So they built a company to try and change that and sell glasses online and also through offline channels. So we'll talk about those guys a little bit. If you happen to be in Boston or New York City, you could even go and visit the flagship store. They have their own real world flagship store selling product directly to customers. The fourth company I'm going to talk about a little bit is a company called Bonobos. This is a little bit later than diapers.com and a little bit earlier than warbyparker.com. It was founded in 2007 by two gentlemen coming out of the Stanford Business School. The current CEO is still there, Andy Dunn, and their idea was to sell men, for all the guys out there, fashion items at great prices, and also fashion items that really sort of fit you and gave a bit of a feel and trim than other products. The final and fifth company on this slide is Citrus Lane, founded by a friend of mine, Mariah out there in California, really addressing the problem that some of us have if we have children, how to get the best toys, the best stuff, the best kind of gifts for our kids on an ongoing basis. Basically to discover new products that are relevant to households that have small children. And if we don't have small children ourselves, we might like to gift that to somebody else. So those are the five companies that we're going to sort of look under the hood to try and understand how all of this works. So I'm just going to explain a little bit about what we do as academics when we examine these things, and what I've done with my friends and co-authors. Is we're able to get the sales data from these companies and match it up with other data, mainly provided by the US government, about the kinds of people that live in different areas of the United States. And then understand how the characteristics of the physical environment affect the sales of a virtual world company. So why is it that one zip code has very, very low sales of diapers.com products and another one may have thousands and thousands of customers? Those are the things we were trying to understand and that was the data that we used to do it. So the next thing I'm going to do is I'm going to take us through three principles that I discovered with my coauthors during this research about how the online world and the offline world interact. If you're interested in reading this, the article's called What Matters Most in Internet Retailing, it was published by the Sloan Management Group. So the first thing that we found kind of summarizes some of the other things that we talked about earlier, is that when an online business opens up, it changes the cost benefit for the shopper in particular locations. So if I'm living in Philadelphia 19123, which is where I do live, and I have access to certain kinds of stores to buy clothing. And let's imagine the closest one is two or three miles away, when an Internet store opens up. Let's call it jeans.com. That changes the relative attractiveness of shopping online versus offline. So that's the first thing, when an Internet company comes into play, it changes the relative cost and benefit of shopping online versus offline. So the second principle that I discovered, primarily in the beginning looking at sales from netgrocer.com starting in 1997, is that the way e-commerce companies developed their sales is very structured and very, very predictable. And so what I'm going to show you now is a very interesting graphic that starts out way back in May, 1997. And there's a picture of the US, you can see there on the screen, United States. And you can see the areas that are shaded. Those shaded areas, areas with somebody in that location, in that zip code, of which there are more than 30,000 in the United States, had actually placed an order at the website netgrocer.com. Way, way back in the dark ages of Internet retailing. Now what's interesting is we fast-forward through the slides, what we can see is that by the time we got about 42 months out, or about 3.5 years out, netgrocer.com was selling products in more than 18,000 zip codes. That's why the whole map now looks rather dark and rather shaded throughout the entire United States. So an interesting question then arises. How did this business grow? How did it go from 34 zip codes in 1997 to 3.5 years later in 2001 selling into more than 18,000 zip codes? What did that look like? And if we were to sit there in January 2001 and say to ourselves, in February 2001, where are the new customers going to come from? Are they just going to pop up randomly in the United States, or will there be some special structure? Now if you're guessing that there's some special structure, you'd be right. And let me now show you the next series of slides that illustrates that. Now what I've done in this slide is I've separated the East Coast and the West Coast of the United States. So primarily around California and primarily around the New York City area, and what you can see as the slide rolls through, is that sales tend to pop up in areas around that had existing sales previously. So you tend to get more new customers showing up in locations that are close to locations that had customers before. Now why is that? Well, there's really two reasons. The first reason is perhaps, I buy something and I live in zip code 19123. And when I'm at my local cafe, or when I'm in my apartment building, I tell one of my neighbors, hey, you really should try soap.com. And so there's a conversation or a word of mouth that takes place. The second thing that goes on that's also very, very interesting is it could be the case that customers are observing the behavior of other customers. You might have asked yourself the question, why is that e-commerce companies have really, really ugly packaging sometimes? Why does our friend Chris at soap.com ship us stuff in a box that's green and brown and orange, and all kinds of colors that if you were wearing them, you would look rather, we'd say clownish? Well, the reason that's done is because the box really kind of pops, and it really stands out. So when my order from soap.com comes to the office at the Wharton School and it sits there in the office, anyone coming into the suite is going to see it. And so that's why you see that very, very structured pattern. There's observation and is also word of mouth. So keep those two things in mind, because when we get to our next module and we're going to talk about how to grow the customer base and how to find those lead users, thinking about leveraging word of mouth and leveraging social observation, is going to be very, very important. Now let's move on to Principle number 3. So Principle number 3 is sort of a further look under the hood of Principle number 2. So you might have wondered, and you can go to the website doppleganger.com, who your celebrity doppleganger is. That is who's your celebrity look-alike? If you were to walk down the street, would people ping you as George Clooney, Brad Pitt, Jessica Biel? Who knows? So also you might imagine for locations, there's the same kind of concept. So perhaps Pennsylvania 19104 is almost the same as another zip code in some other place in Texas. And so we wanted to examine this idea of whether or not similar locations that were far away from each other will also start to buy things online at roughly the same time. So the original idea for this actually comes from a very, very interesting study conducted way back on the 1970s by a sociologist. And what he did is he kind of changed our idea of distance from just purely physical distance to also thinking about social distance. What do I mean by social distance? So in the original study, Professor Fisher showed that if you lived in Chicago, there was a higher chance that you would meet or randomly interact with somebody from Los Angeles, which is very, very far from Chicago. Than there would be that you would interact with somebody from Springfield, Illinois, which is much closer. So even though Chicago and Springfield are close together, in many other ways they're quite different. The kinds of people that live there, the way they think, the stuff they like is very different. Even though Chicago and Los Angeles are very, very far apart in terms of distance, there could be a lot more similarity in terms of the taste of people that live there. They both like big cities, they both like to consume certain kinds of goods and services. So continuing with our discussion of Principle number 3, what we found was very, very interesting. When we looked at the sales in the Internet retailer, initially those sales started to take off in larger cities and spread, like the pattern that I showed you earlier, spread through the notion of proximity. Meaning, people living close together were either telling each other about the good or the service or perhaps copying each other from seeing discarded boxes and so on. So in the beginning of the sales process, those sales started to spread out by proximity from one customer to another. But what happened over time is very, very interesting. Sales in those key locations started to, for want of a better term, taper out and reach a steady state, and then pick up in other locations that were farther apart from each other, but yet shared very important characteristics. So you might have two locations in different parts of the United States, maybe even 1,000 miles apart, but the basic profile in terms of the age, income, occupation, education level and so on and also of the retail environment will be very, very similar. Those are those doppelganger locations that I talked about. And because those locations are very similar in terms of who lives there and the opportunities for shopping offline, they tend to migrate online at roughly the same rate. What we did when we discovered this in the beginning through proximity, over time more through similarity of locations, that we were able to develop something we called the long tail plot, again borrowing from our friend, Chris Anderson, but this time a long tail plot over location. So if you look at the slide, what you'll see is you'll see, on the x axis, different locations in the United States and the y axis is the level of sales. So the sales start very, very high in the best location, and then they taper down to the lower locations. But those lower locations that generate small sales levels individually are still very, very important for an e-commerce company. That's the key takeaway. You can't just survive by hitting the big markets. You have to also hit the many smaller markets that collectively add up to a lot. It's the same idea as Chris Anderson's long tail, but this time if you think back to that previous discussion, this time the x axis is about rotation as opposed to being about products. [MUSIC]
