Will patience pay off?

Sky Dayton, who started EarthLink, now the nation's No. 2 Internet service provider, may again be in the right place at the right time.

In late 2001, Dayton founded Boingo Wireless to sell Wi-Fi access to subscribers around the United States. Instead of building a new network, Dayton's big idea was to patch together a Wi-Fi network out of existing providers--the same idea that he used to create EarthLink.

Since then, Wi-Fi card prices have fallen, and the technology is increasingly supported by notebook computer markers. What's more, the number of installations of hot spots--public areas where broadband access is available through Wi-Fi networks--is increasing.

Taken together, all of this is expected to help support an easy-to-understand hot spot business model. However, the service industry still has to continue to expand the scope of wireless networks with improved seamless roaming for the market to grow.

But as the numbers grow, so to do the expectations. CNET News.com caught up with Dayton for his perspective on the future of Wi-Fi technology and wireless computing.

Q: How will Wi-Fi affect people's perception of wireless?
A: Wi-Fi is introducing a lot of people to the concept of wireless data, and they are going to think, "Hey, I want 3G in my car, or I want to use it where Wi-Fi isn't available." At the same time, it's increasing the range and variety of devices to connect to networks. So, it's a whole new range of applications that we haven't thought of. An iPod is a natural. You can download music before you jump on a plane. As chip prices come down, Wi-Fi can be built into everything--digital cameras, Game Boys and cell phones.

Where does Boingo Wireless fit into the hot spot scene?
Boingo is really an enabler to let major brand players have a viable Wi-Fi offering. So, we work with EarthLink and T-Mobile to allow their customers to get access to these networks easily.

How does the hot spot market compare with the cellular market?
Wi-Fi turns the traditional wireless model on its ear because the carriers don't need to worry about whether the user is going to need a device. The user already has the device that connects to the network before the network is even built. People are getting those devices at home and to use in the office, and I think it is safe to say that every broadband endpoint is going to have a Wi-Fi connection on the end of it. The last hundred feet of the Internet for users in homes and offices is going to be Wi-Fi.

What impact will Wi-Fi have on next-generation cellular technology?
Wi-Fi certainly challenges some assumptions that people have about 3G. It doesn't replace 3G by any means. In fact, 3G is better off because of Wi-Fi. It solves some of the problems you have with 3G in that it's hard to deal with high demand in a congested area with a wide-area network. You can't deliver the kind of high bandwidth over extended ranges. You need Wi-Fi for that.

The more carriers look at it, the more they're going to realize that this is going to be a good way to offload demand to a much lower cost system that in many cases has more bandwidth. The two puzzle pieces have to fit together, Wi-Fi and 3G, and the first companies to figure out how to deliver both are just going to explode with success.

Everyone likes to pit the two against each other.
Yeah, it makes for interesting debate, but it's really not one versus another. It's a complementary situation. There are much more controversial issues, such as voice over Wi-Fi.

One beef that analysts have with hot spot service is the absence of any model that makes money.

You have to look at the business model, not whether or not anyone is making any money today.

We're in a development stage of a new industry. You have to look at the business model, not whether or not anyone is making any money today. The business model is very easy to understand, and it's driven by low cost.

Do the math for me.
It costs $150 a month in all to provide the service, or $5 a day. Can you generate $5 per day per location in order to break even? Let's say you have 300 customers walking into your coffeehouse per day, and you have a 2 percent conversion--which is low--so that's about six people a day. Boingo pays the hot spot operator approximately $2 per connection, and that's what the cafe owner would get. So six customers a day times $2 is $12 per day. That's $360 a month on a $150 cost, which is a 60 percent profit margin. And that's before we have the kind of ubiquity in Wi-Fi devices that we're going to have.

So when you walk into a Starbucks three years from now, what will you see?
When you walk into a Starbucks three years from now you're going to have multiple Wi-Fi devices on your person. You're going to have your laptop, your PDA, your phone, your MP3 player--who knows what else? And all of those will be aware and be able to connect and download and update, etc. The business model is clear but it's early. I think it's totally reasonable for people to have questions.

What are the big challenges for service providers like Boingo?
One is the lack of ubiquity. (Wi-Fi) not yet in all the places you want to use it. Ubiquity isn't the right word because it's not a wide-area network. It's for places where people are concentrated and want access. It's got to be in airports, hotels, cafes, convention centers, campuses. But it doesn't need to be out on the highway, and it doesn't need to be on the golf course.

So the idea of a nationwide network doesn't really make sense?
It's never been about that. It's about pinpoints of access. By the end of this year or early next year, we're going to go from about 5,000 total commercial hot spots in the U.S. to about 10,000. We're going to be close to having it in all the right places, which means all the airports, all the business-class hotels, all the convention centers and ubiquitous brand cafes and places like that will follow. Those aren't as critical. That's when you start to feel like it's in all the places that you need to depend upon it.

Is that the big barrier you need to hurdle?
That's the first barrier. The second is fragmentation. When you walk down the street in a city with a Wi-Fi device you're likely to run across several hot spots and in order to access each of them you're going to need several different accounts. That's not convenient and is not going to work. There needs to be seamless roaming across the industry and it's even more important to Wi-Fi than it was to cellular because Wi-Fi has a smaller range and roaming is part of the very fabric of the business.

Until we have universal roaming, which is what Boingo is trying to achieve, we'll be holding the entire industry back. The third barrier is ease of use. We're making great strides with that.

How do you see pricing models change as more people access Wi-Fi over a variety of devices?

Ultimately, you'll get to the point where a $50 a month plan will give you all you can eat over Wi-Fi and 3G networks on multiple devices.

Ultimately, you'll get to the point where a $50 a month plan will give you all you can eat over Wi-Fi and 3G networks on multiple devices. But we haven't found the magic price point like we did in the ISP business at $19.95 per month. The average revenue per user is going to be higher with wireless than it has traditionally been across the board because it's broadband. It's neither narrowband nor wired, which are two things you have in dial-up.

We've already seen one carrier, Verizon Communications, bundle Wi-Fi with another service. Is more of that in the future?
Absolutely, and both will be successful. We like to talk in simplicities, black and white, but it's rarely the case.

A lot of this business seems to be based on partnerships, but that puts companies in the awkward position of being both a competitor and a partner. Is that just the way it's going to be?
It's healthy for an industry in the early days to have channel conflicts because that implies that you have channels and demand. Wayport was really smart; they recognized that early on and they adopted a strategy of not only selling directly to enterprises and end users but also selling to Boingo, Verizon and anybody that wants to come and buy access from them on a wholesale basis. And all it does is help fill up their pipe. The hot spot layer is an inherently fixed cost business, so the more traffic you drive against it, the higher the profitability.

This is not always the case in places where the capacity is scarce, like the Bell world, where the underlying physical asset is a monopoly. But that's not a concern with hot spots. Like the Internet, companies thrive through partnerships and being really focused on what they do and partnering.

And the same applies with Wi-Fi.
Everyone is dependent on someone upstream in the value chain. The raw materials of Wi-Fi are the venues; the operators depend upon the venues; the roaming aggregators depend upon the hot spot providers; and the brands depend upon the roaming aggregators.

That makes specialization very important then.
Exactly. With specialization you get the edge. The little details are what will determine success and you can only understand those by living and breathing the experience. You can't get that by generalizing and spreading yourself too thin.