Nearly a quarter of commerce now occurs online in the United States. Yet, according to a recent research paper by David R. Bell and Wharton doctoral candidate Jae Young Lee, “location, location, location” is as relevant in the world of e-commerce as it is in the physical world. In the paper, “Neighborhood Social Capital and Social Learning for Experience Attributes of Products,” Bell and Lee offer surprising insights about how our offline behavior — including where we live –influences our online behavior.

This research served as a partial basis for Bell’s new book, Location Is (Still) Everything: The Surprising Influence of the Real World on How We Search, Shop, and Sell in the Virtual One. In the book, Bell reveals how our physical world shapes the ways that we use the Internet and explains what online sellers need to know to understand the conditions in which their customers operate.

In the following interview with Knowledge at Wharton, Bell discusses the takeaways from the research paper and poses questions for online sellers to consider: Why do you have customers in some locations and not others? How does the physical world affect what’s happening in the virtual world? How can you inspire your current customers to advocate for your products and services in the real world?

Edited excerpts from the conversation follow.

On ‘neighborhood social capital’:

This research is called “Neighborhood Social Capital and Social Learning for Experience Attributes of Products.” So, right there in the title are some of the key ingredients that we want to think about.

The first is neighborhood social capital, meaning the extent to which people trust and know each other, and interact with each other offline. Social learning means that information is transferred from an existing customer to a potential customer. And the experience attributes [of] products [mean it is] really hard to know what they’re going to be like until you try them. We’re thinking about fashion apparel, eyewear, those kinds of things where you would really like to have a visceral experience. This research is about how to figure out where those customers live, and who is most likely to talk about the product and therefore generate new sales, particularly for products where some explanation is required.

On the key takeaways of the research:

There are two really important takeaways here for anyone who is in the online world selling things through the Internet. The first is that it’s very critical to understand what the offline environment looks like. Why is it that you have some customers in one location and no customers in another? What is it about the physical world that is affecting what is going on in the virtual world? That’s the first thing.

The second thing that this research emphasizes is that existing customers are often the most powerful source of new customers for a firm — the extent to which they advocate to [others], the [degree to which] the product they’re using is socially visible, and the extent to which they talk about it with friends, family and others who they come into contact with locally.

“There are certain things that online environment does for you — it makes fulfillment easy, it makes it easy to scale. But there are other critical things that offline does for you.”

On his most surprising conclusions:

With any piece of research, it’s always nice to find something that’s a little bit surprising or counter-intuitive. And there were really three things here. The first thing was that this social and learning effect was really quite large. Up to half of the new customers that came into the firm that we studied — [men’s clothing site] — got there because the information was shared from one customer to another in an offline environment. That was the first thing.

The second thing was that this effect became more important over time. People who were the first customers into the firm — maybe they’re going off their own information or … they are fairly risk-seeking people. People who came later on tend to require a little bit more hand holding from others.

The third thing is that this notion of social capital — or the extent to which people trust and interact with each other in a local neighborhood — doesn’t generate sales per se, but it makes information transfer more efficient. So, if bad information is being spread — i.e. “This company’s no good” — it’s going to slow down sales. In this case, fortunately for Bonobos, what was being transmitted was positive and therefore sales sped up.

On the practical implications of the research:

The first thing that’s really important for any commerce business to understand is that the offline environment is going to explain a lot about the success in online sales. The firm really needs to think about what kind of locations are going to be most fruitful and why — I mean, [leadership needs to] go into those locations and see customers.

Secondly, … some products are naturally socially visible — if I’m wearing glasses or I’m wearing a pair of pants, people might ask me about them or I might be able to generate a conversation. For products that are not naturally socially visible, then maybe there is a way to make it socially visible by having it shipped to the office rather than the home, for example.

Let’s give a practical example that really ties all of this together. The firm that we looked at — — is a men’s fashion retailer. What we found was, in locations where customers were more apt to talk to each other and trust each other, there was a greater sales diffusion online. The target customer in this case is a male, aged 25-45, who is somewhat fashion-forward. It turns out that a good proxy for where those males are congregating is the number of bars and liquor stores per capita in a location. We had some sociological theory that told us about interaction and then we were able to go to public data and find a variable that was actually a pretty good proxy. So, it’s tying those things together, thinking about the theory and then trying to find a proxy with real data.

On possible strategies based on the research:

One strategy that a new commerce firm can pursue, which is really just an outcrop of what they’re already doing, is to append the zip code data that they collect naturally through the sales process with other data about what is going on in that zip code. This can come from the U.S. Census or for other third-party suppliers.

For example, if I’m Bonobos I would like to know how many fashion apparel stores there were in a location. I’d like to know what the population density was, what the age distribution was, and about the presence or absence of a college campus. Those things will tell me quite a lot about why I might have high sales at one location and low sales in another. So, it’s really just dependent on what has already been done. That’s the first thing that a new commerce company should be thinking about.

“Social capital — or the extent to which people trust and interact with each other in a local neighborhood — doesn’t generate sales per se, but it makes information transfer more efficient.”

On taking the conclusions beyond fashion:

This is really a very general conclusion. Even though there are a number of companies out there now selling things online — through the virtual world, if you will — all of those companies are selling to customers who still live in the real world. So, understanding the real world circumstance of your customer — what other offline options they have, who they live next to, how they interact with the local environment — is a really critical ingredient, perhaps the most critical ingredient, for understanding whether or not they’re going to shop online or offline, how they’re going to search, and also with whom they’re going to interact and spread the good news about your product.

On how the research relates to recent news:

One thing that’s getting a lot of press in the news right now — in fact, I heard a great buzz word and being in marketing, we love buzz words — is this notion of “omni-channel” i.e. online and offline being integrated seamlessly. What we’ve seen is there are firms who came out very strongly early on — perhaps even Bonobos, which was the subject in this study in 2007 when they first started saying, “Why would anyone ever bother to operate offline when you can sell through this wondrous thing called the Internet?”

But what everybody is realizing now, whether you are Rent the Runway or Birchbox or even Amazon, in addition to Warby Parker and and many others, is that you really need to operate both online and offline. There are certain things that the online environment does for you — it makes fulfillment easy, it makes it easy to scale. But there are other critical things that offline does for you, as well. It makes it much easier to relay information to other people. It also makes it much easier to build credibility about your brand if you really have a physical world presence.

We’re seeing more and more of this discussion now — what it means to be omni-channel — not just present in one world, but present in both.

On how the research dispels misperceptions:

This study dispels a couple of misconceptions. The first misconception that I’ve talked about already a little bit is the fact that the online world is somehow disconnected from the offline. The offline world explains a lot of what’s going on in the online world. And also, the offline world is actually pretty easy to understand. There has been a whole history of very, very rich data being collected about who is living where, what are they doing, the level of commercial activity and so on.

The second thing — and we’ve heard this for years and years in commerce — is that location is very important. But we tend to think about the location of the firm — where do we put the store? Where do we operate the factory? Which country should we be in? But what we’re finding is that it’s still about location, but this time it’s about the location of the customer. Where is that customer and with whom does that customer also live? That’s what’s really important in the world of e-commerce.

“The offline environment is going to explain a lot about the success in online sales.”

On how the research is unique:

One really novel aspect of this research is that it is really combing ideas from two different fields and data, in fact, from two different fields. In this case, we’re talking about some ideas that come from sociology — why it is that people trust each other and interact with each other — and also some ideas from marketing about how information gets communicated, and how people relay ideas from one person to the next.

It’s also combining data from two disparate sources that are not often used. The first source of data is the Social Capital Community Benchmark Survey. I know that’s a mouthful. But it’s coming from the [Harvard] Kennedy School, collected by Robert Putnam, who was involved in writing the book Bowling Alone, which is about the collapse and revival of social capital in America.

And then secondly, [we used] data from the firm itself —, founded by Andy Dunn. What we were able to do is merge theories from sociology and marketing, and also merge data collected for another purpose, with real data coming from a prominent e-commerce firm.

On what’s next:

This is a very rich area, this whole online, offline, omni-channel interaction. The next thing I’d like to look at is how customers who are exposed to the firm in one world — let’s say the offline world — might continue to interact with the firm through the online world. A lot is being written right now about pop-up stores. [For example,] Warby Parker has a bus. “We’re heading around the U.S.” and so on.

I want to give a shout-out to Lawrence Lenihan, one of our alums who runs FirstMark Capital in New York. He spoke to my class and he gave a very, very interesting analogy. He said that, for example, when I interact with somebody — let’s call him “Joe” — and it’s over e-mail, it’s a very low-energy interaction. We don’t know each other. And then Joe and I get on the phone and Joe hears my accent, I hear his Long Island accent — and our energy level goes up. And then we go out and we have a couple of beers and we have dinner and we really get to know each other. So, then the next time an e-mail comes out, the energy level has now been elevated.

And so it is, in some sense, with all online or offline. We’re going to look at the customers that were acquired by Warby Parker for eyewear and Bonobos for fashion clothing. If the first experience was offline and it was a good experience, are the firms able then to subsequently serve these customers online? And there are great operational advantages from doing so because it’s much, much cheaper to fulfill that customer out of a warehouse than to always having him come into a retail store. That’s really where I want to go next.