|This is Chapter 9 of Rory Conaway’s series, “Tales From The Towers”, about working with municipalities when you deploy a large scale Wi-Fi network. If you have not yet done so, read Chapter 8: Half of something is better than nothing, but why settle?|
Now let’s talk about the real-world in deploying this type of system.
Realistically, light poles or traffic lights are the obvious deployment locations. They are usually spaced pretty optimally. The problem is they are either owned or operated by the local municipality or the local power company which in many cases, is overseen by a local governing board. Either way, the key word here is government.
I have dealt with municipalities that are both easy to work with and others that are very difficult to work with. It always comes down to the individuals within the departments. However, it also always takes a consensus of staff or departments to get a project through. The ability to get a project passed goes down as more people are required to make a decision and the more money the department has to expend. So getting approval to use municipal facilities generally takes a pretty long time, especially if involves the budgetary cycle.
Now you have to throw in the legalities of a deployment. You need some type of site lease for any facilities you are mounting equipment on. In addition, you might also need insurance in case something gets damaged. We have been required to carry a $2 million dollar policy for some locations. If you have to deal with property owners for buildings, the contracts get more complicated as there is the issue of sale of the building, access, etc. Even though technically Triadland looks pretty easy, the devil is in the details.
When you go back to build financials for a deployment and you start adding in all the back end costs, you can see that unless you are expecting a lot of paying users, in the thousands, a multi-million dollar capital expenditure (Capex) is going to kill the project unless you accurately predict the costs. If you are utilizing a lot of vertical assets, such as some of the models using 60 access points (APs) per square mile, and you have to pay for the rights to each of these poles with fees such as $10 per month per pole, then the monthly cost goes way up. Of course knowing you need 60 APs up front instead of thinking you need 30 APs and then later finding out you really needed 60 to begin with is far worse. The days of building a system and the users will come went away with the cell phone companies providing better bandwidth options. Although even Apple knows that WiFi can provide the bandwidth it needs for video applications, the cost and difficulty of managing a few towers versus thousands of APs makes it a no-brainer for cellular providers.
Although some of the cable companies have moved into the WiFi area with hotspots to enhance their security, it’s not really a national standard for a business model. There is even discussion of cellular companies paying WiFi providers for access to their network. However, it’s evident that although WiFi has high bandwidth capacity, defining the target client is key to profitability unless Verizon throws some petty cash your way.
As with any good business plan, in complete contrast to former companies that have failed in high profile examples, the key is keeping Capex down, keeping monthly expenses down, providing a good quality service to retain and add clients, defining a realistic revenue stream, defining system expectations, and making sure you have control of as many facets of the product environment as possible. Sometimes this results in compromises in technical capabilities to balance out profitability and success. Building a system that costs $10,000 per square mile but generates $3000 per month in revenue with a $1000 per month of expenses is far better than building a $100,000 per square mile system that generates $6000 per month in revenue and costs $5,000 per month to support (keep in mind the cost of money needed to build out the system and manufacturer warranty costs) is not a better deal. There is nothing wrong with saying we can’t give 100% indoor coverage if 40% makes it profitable and 80% indoor coverage makes it unprofitable. If the system fails financially, it’s irrelevant what percentage gets covered.
I bring these examples up to show that it takes different designs for success in different applications. Although the industry idea of a municipal wireless system has traditionally been stick APs up every few hundred feet and build a system that is everything to everybody, the best design is the one that is self-sustaining financially which motivates the operator to keep it running.
Municipal wireless networks by definition used to mean 100% coverage everywhere, similar to cell phones, with a universal product that was everything to everybody. We need to redefine it so that it means any 802.11 service protocol deployed in a Wide Area model to provide bandwidth using a unique methodology that can’t be provided by existing cellular or satellite models. Start there and we have a whole new set of opportunities. Throw in 802.11N, MIMO, new 3.65GHz and 900MHz equipment, very low cost wireless equipment, and the possibilities are endless.
Instead of going to a municipality and asking them to be an anchor tenant, why not ask them to pay $10 per month to place a camera on your network or even better, you own the camera and lease viewing time to the city? It’s far easier for the police department and water department to find the funding to pay $100 a month for access to your camera than to Capex $1000 for a camera. Not only is the source of the funding different meaning different rules for purchasing, it also relieves them of hardware obsolescence, technical support costs, and internal training. Expand this idea out to other areas as well.
Many governmental departments are paying $40-$60 per month for cellular data cards. I can tell you for a fact that they are a mixed blessing. The lower bandwidth doesn’t work well for video uplinks. There are dead spots or areas where many of the systems slow down to lower rates and in many smaller cities, they are limited to 100K or less systems. Yes I know LTE and Wimax are here in many cities but try uploading a 3-6Mbps video stream and tell me how that works. Put up a system that can provide better outdoor coverage and they would gladly pay you instead of Verizon or Sprint. Assuming you have outdoor coverage that works in vehicles as well, you don’t need laptop coverage everywhere because in reality, the city vehicle is a mobile hotspot itself. Use 2 APs in the car, $200 instead of $100, and you have a mobile hotspot with about 200 to 400 feet (60 to 120 meters) of coverage around the car. For improved performance in a mobile environment, a dedicated router product such as a Peplink Mobile Max or CarFi, or something similar might be better. I like the Peplink Mobile Max product because it has a load-balancing hardware VPN tunneling feature coming out for secured communications. In mission critical environments, I’ve even looked at using 2.4GHz and 900MHz in the same car simultaneously.
I know that it’s hard to get rid of the old ways. There are lots of manufacturers still building the same equipment they sold 5 years ago with minor changes to support 802.11N. Unfortunately, not even all of them have the ability or design to upgrade to 802.11N. The AP manufacturers need to figure out how to build a product that can make a deployment profitable if they want to get back into that market. If municipalities are the only clients, there is a limited market for too many manufacturers. What is needed is a model where venture capitalists, many of whom have already been burned, see a business model that is not based on advertising, but based on reasonable costs and has solid income potential. It can’t be based on the idea that government officials are needed to make a business viable. We as consultants need to suggest that manufacturers get entry level prices way down, find additional sources of WISP revenue, and create an expectancy of profitability to drive the market. The manufacturer who does that will own the market.