ADI interview - provider to large Apartment buildings in New York

nPost.com: I am here with Marc Josephson of ADI, Advanced Digital Internet, thank you very much for meeting with me today. What exactly is ADI?

Marc: ADI was started a little over two years ago to provide very high speed Internet access to large residential building in New York City. We currently have 70 building in New York City, and have recently launched in Philadelphia. Our goal is to bring a high speed T1 or T3 lines into the basement of a building, install a router, and run Ethernet cables up to each floor. We initially started wiring each apartment via Ethernet cables, but we found that it was too expensive and so we started providing wireless nodes based on 802.11. Our customers have wireless access from their apartments, and this solutions saved us the cost of doing the horizontal wiring. The users also now have the flexibility provided by a wireless solution, and it is also much more aesthetically pleasing in that we have wired some very old apartment buildings and they did not want wires stapled to the ceilings of their hallways.

The advantage of the Wireless LAN is that it has an 11-megabit capacity, which is more than enough to handle residential Internet access. It has been a good foundation for us, and we are now beginning to upgrade to 802.11a, which is 54 megabits. With that we will be able to start delivering video and other high bandwidth applications.

nPost.com: How do you market these services? Is it directly to the building owner, manager, or the end user?

Marc: It is a combination. First of all, in order to build the infrastructure such as installing the wires, transmitters, etc. we have to have approval from the landlord. The first thing we do is we meet with the building owner, and determine if they are interested in letting us provide service. If they are, we then perform a site survey to determine the best way to approach that building. Doing this we are able to determine the cost of wiring each building, and we will then go back to the landlord and reach an agreement of whether or not to move forward. If both parties are interested we will then enter into a Building Services Agreement, which gives us the permission to install wires, routers and transmitters.

What we look for is the active cooperation of the building owner and management. While we sell directly to the end user we don't want to barrage people, stuff mailboxes, etc. We have found the most success if the building owner or manager sends a letter to the tenets mentioning our services, and providing ADI contact information. We will also arrange to do a lobby presentation, where we typically have a couple laptops with a wireless connection and streaming media presentation for people to take a look at when they arrive home from work.

nPost.com: Do you also provide the wireless connection for their desktop or laptop?

Marc: Absolutely, the great thing is that we started doing this when Windows 95 was the norm, and installing a wireless connection required a trained technician. Now Windows 2000 and Apple's AirPort have this capability bundled in. It is now plug-n-play, and the software installation is quite trivial. At least half of our users self-install and the other half pay us to have a technician come in and make everything work.

nPost.com: It is very capital intensive to wire a building, and you don't start wiring a building with customers already lined up.

Marc: That is an excellent point, and I actually skipped a step. After we figure out the engineering we like to perform a survey of the tenets, whether through the owner or ourselves, to determine the level of interest for our service. If there is a very low level of interest we are able to tell the owner that the deal doesn't make sense and save our capital for buildings that do have the interest. Also, as part of the Building Service Agreement we also give a free network connection to the Super or the doorman.

nPost.com: How many users are required to make a building deployment viable for ADI?

Marc: The rough rule of thumb is that we require about 20 customers in one building to break even. With the communication expense of bringing a T1 or T3 to the building, and we also have the one time expense of the routers, the cables, and the transmitters, which we amortize over a three-year period.

nPost.com: Your wireless solution really enables you to minimize your costs associated with adding additional users.

Marc: After we have been in a building for three to four months penetration really begins to ramp up.

nPost.com: What are you seeing as your average penetration for a New York apartment building?

Marc: Well, we are seeing about a 30% penetration after the first year, and at the end of the second year we are getting closer to 40%.

nPost.com: From an economic standpoint you are limited to apartment building with greater than 100 units.

Marc: Any smaller, and it is really risky for us. We have done a few brownstones that have 10 to 15 apartments and we provide a flat rate for the whole building. That solution has worked well for customers that don't have access to Time Warner Cable or DSL. Each tenet will typically pay $50 per month, which is much more than normal, but at the same time they have access to almost unlimited bandwidth.

nPost.com: What has been your competitive positioning against Time Warner Cable, Excite@Home, and DSL?

Marc: We are looking for building in which there are no alternatives, where we can be the only solution. In building where these are both DSL and Cable Modem we find that we get the highest proportion of users because we guarantee the bandwidth. A user with a Cable Model shares that bandwidth with many others, and DSL is limited from a bandwidth standpoint. At peak periods both of those other technologies slow down because of the number of users that are online. When we sign up a new customer we provide different levels of bandwidth with associated pricing.

nPost.com: How do you ensure that each user has access to their bandwidth level?

Marc: We monitor the bandwidth used, and as soon as a building reaches 75% of their capacity we order more capacity to the building. We have migrated from T1, which have 1.5-megabit capacity to T3, which have 45-megabit capacity.

nPost.com: How is the laying of the T1 and T3 cables facilitated? Does ADI actually lay the cable or do you work with local carriers?

Marc: We use local carriers such a Verizon, ConEdison, and Level Three.

nPost.com: With wireless LANs one of the major issues is security. How is ADI approaching security and what is it doing to ensure it/

Marc: Couple fold. We employ an encryption program, and we have set up a VLAN for each user. The VLAN provides an encrypted tunnel to each user, although a sophisticated hacker could break it, just like they can break into anything. Our system is not perfect, but it is much more secure than other solutions.

nPost.com: Exploring your growth strategy; do you employ feet on the street working with Property Managers, building owners, etc.

Marc: Yes, we have a dedicated sales force. We have also been able to develop lasting relationships with the Property Managers that we have worked with which has led to significant sales via word of mouth. Our direct sales force also comes from the Real Estate industry, and has a lot of contacts and experience in this area.

nPost.com: You mentioned that you also provide a wireless capability to the building Manager, Super or Doorman, how does that then work into your BAM Software Solution?

Marc: BAM or Building Assistant Manager was a software package developed by a Building Superintendent in the Trump Organization. He had been managing high-end properties within that organization and was very highly regarded. He wrote his own software to help manage the properties. The functionality is incredible, but the problem was that this was his first programming job, and he was learning as he went. The modules don't follow a consistent architecture, so we acquired the program from him and have rewritten most of it. He had the insights, and the requirements to write a dynamic product.

nPost.com: He knew his needs.

Marc: Exactly, from that standpoint the software is excellent. One of the modules he had developed for the software is one called Package Tracking. You get a large apartment building and a lot of the packages arrive during the day, when the tenets are off at work. The tenet gets home and there is a different doorman on, and they don't get notified that their package is waiting. We took the functionality he had created for tracking packages and put it into a stand-alone module. We then added a bar code scanner, which logs in the package information to the computer, the Addressee, date, time, etc (from UPS, FedEx). Once a package is logged in the software automatically sends an email to that resident notifying them that they have a package waiting.

nPost.com: The market for high speed Internet access has received a lot of press lately with the bankruptcy filing by Excite@Home and the financial straights of DSL providers. What is it going to take to be successful in this market?

Marc: @Home built a wonderful broadband network. They also figured that they needed to develop a portal so that they could grab the users. Not only would they be the network, but would also be the content play. We also worked with a number of Real Estate owners who convinced us that we needed to do that as well, and we developed one that no one uses. Yahoo! and Excite do a much better job than we ever could. We had to decide which business we wanted to be in, was it providing Internet access or being a portal? We decided that our core strength was obviously Internet access. With the collapse of Internet advertising that model is no longer viable.

nPost.com: You decided to focus on your core competency; Internet access, how did that help you?

Marc: We have been able to do something, which is extremely difficult, which is to put together reliable high speed Internet access and be able to make a profit. Which is why we are one of the last high speed access provider that is not part of a cable company.

nPost.com: Many providers have had major issues in solving the issue of the "last mile". What were the factors that impacted you in driving a profitable last mile solution?

Marc: How much time do we have? It requires solving multiple problems simultaneously. We took a very strategic look at the business and tried to determine the rates of decreasing cost in different components. If you came up with a solution for a problem three years ago you will have a much different structure today. Going forward it is also going to be very different. The trick is to design an architecture that can adapt with the structural changes over time.

nPost.com: The network has to incorporate as much flexibility as possible, because you can't determine where the changes will be.

Marc: Well, were able to accurately guess where the changes were going to be made. One key was that we took a look at where the greatest improvements in hardware price performance has been, consumer electronics. As PCs became consumer electronics they began following Moore's Law. Every year, they became better, cheaper, and faster than the year before. Communications equipment has improved, but nowhere near as fast. The phone business has not really seen that dramatic type of improvements, which is one reason it is so difficult to compete in the phone business. Take a look at LAN equipment; Ethernet routers, hubs and switches price points. You can now buy an Ethernet switch for about $50, which is as powerful as one that cost $1,000 three years ago. We have tried to build our infrastructure on Ethernet equipment versus communication equipment. So that over time our components are become cheaper and more powerful.

nPost.com: What have been some of you keys to success in starting ADI?

Marc: Very simple; hire good people and be very frugal. I tell my people to be very careful with the money that they spend, because you can only spend it once.

nPost.com: What do you see as a key to success in this market? Is it to continue and grow your base of NYC customers and the number of service offerings your provide, or is it to expand your model nationally?

Marc: It is both. On the residential side we are finding the added values and services are getting a lot more attention than they used to. Although, they aren't a large moneymaker yet, but we believe they will be. Just Internet access is enough to provide a viable business, and if we can make a profit providing Internet access all these other services are additional gravy. Whereas a lot of our competitors constructed business models that required them to sell a bundle of service in order to make money, which clearly didn't work. With a lot of these bundled packages some of the products are nice, and some are so-so.

nPost.com: And consumers want the best in every category.

Marc: They are willing to sacrifice a little if the product or service is really convenient or they are getting something special.

nPost.com: Do you foresee providing services such as Movies on demand, Internet telephony, etc. through your network?

Marc: For the last four years I have been saying that voice over IP is right around the corner. I just haven't got the right year and or decade as of yet. What has happened is that the long distance capacity issues have forced down the cost so dramatically that there was never an economic break-even point for voice over IP. The technology works, it just isn't economically viable.

nPost.com: And video?

Marc: We are specifically planning to do video, and we have actually developed a technology for video on demand that works. You can actually see the demo behind me (points to server and video equipment behind him in conference room) that we have set up for a demonstration that we conducted yesterday.

nPost.com: Another important factor is that you provide a service that users actually pay for. Having just resigned from Intellispace what are your plans moving forward both with ADI and other pursuits?

Marc: Well one of the key reasons I resigned from Intellispace was that I hired a bunch of really good managers who were really running the day-to-day operations. With their model it is now much more of an evolutionary approach to the business. It no longer requires the dramatic change that I could bring to it.

Also, the events of September 11th, 2001 really got to me, and I decided to devote a significant amount of my time in Pro Bono work for the City (NYC) and the Federal Gov't. The goal here would be to make the telecommunications infrastructure more reliable. I have been working with the Mayor's Office, and I am meeting with some Senator's and Congressmen next week, and I have been meeting with other Business and Community leaders in the New York area to say how do we facilitate a more robust communications infrastructure