Can public Wi-Fi services be profitable? This is a question being debated across North America right now, with opinions differing widely. Tatara Systems, which began considering this question more than a year ago, decided to look for an answer by creating a basic business model that focuses on the short-term opportunity for public Wi-Fi services.

 A Look At Wi-Fi Operational Costs: Airport Case Study
Consider the Wi-Fi hotspot buildout costs experienced by a wholesale Wi-Fi network operator -- that is, an operator that owns public Wi-Fi hotspot networks and that allows other retail service providers’ end users to use the networks through roaming agreements.

For this case study, we’ll look at a large airport such as John F. Kennedy International. Switches, routers, T1 multiplexing hardware and wiring are needed to complete the network buildout. For the model, we’ve estimated that 75 access points, five switches and a single large router plus T1 mux will be needed to cover the entire airport. The total costs for the network build-out -- which we’ve estimated at $230,000 -- also include wiring and labor. If these costs are amortized across a five-year period at a seven percent cost of capital, the annual amortization is approximately $55,000.

The other major cost item is backhaul, which will be driven by the average data volume per session and the distribution of sessions, since backhaul will need to be scaled to meet the peak usage level. We’ll estimate backhaul costs based on the following assumptions:

• Average session volume = 30 MB

• Average session duration = 30 minutes

• Peak day traffic = 22 percent

• Peak window duration = 180 minutes

• Peak window usage = 70 percent

• Target peak backhaul utilization = 60 percent

• Cost of a T1/T3 = $400/$4,000 per month

We scale the backhaul so that even at peak times, each user is receiving adequate bandwidth -- with the 40 percent excess capacity available to support traffic ‘bursts.’ Finally, we’ll add an additional $24,000/year for network maintenance -- assumed to be low based on the use of high-end equipment.

The chart below illustrates the results of our model for the values indicated. Two data points, indicating the number of sessions that would be generated if one percent and two percent respectively of the enplaned passengers in an average month were to use the network, have been overlaid on the chart for perspective.

As we can see from the graph, Wi-Fi sessions with high (30 MB) data volumes can be delivered at a low cost as long as the network is able to generate reasonable usage volumes. At low volumes -- say less than 2,000 sessions per month -- the costs become fairly high.

WHOLESALE NETWORK OPERATOR PERSPECTIVE
By expanding the operational cost model above, one can begin to determine the usage volumes and pricing needed to make a public Wi-Fi network profitable. Let’s look at some ranges for other costs, including payments to the property owner, sales and settlements that the wholesale network operator will incur.

Payments made to the property owner will be a key determinant of the profitability of the network. These fees might be structured as either a percentage of revenue or as a fixed fee per session, or the property owner might instead require a fixed annual payment. Let’s assume the latter is the case for a top 20 airport. If the annual payment is $100,000 and the airport has 18 million enplaned passengers per year (approximately average for the top 20 airports in the US) and one percent of the enplaned passengers use the network, the fee equates to $0.55 per session. If three percent of the enplaned passengers log on, then the fee drops to $0.19 per session.

Let’s estimate selling costs -- signing up retail service provider partners -- at $100,000 per year. Finally, we’ll estimate that billing and settlement costs are outsourced to a clearinghouse on a per-transaction pricing model. The transaction volumes at one percent and three percent of traffic are 180,000 and 540,000, respectively. A reasonable range of fees might be from $0.10 per transaction to $0.50 per transaction -- with the price declining with volume.

Combining these estimates with data from the case study, the table below summarizes the total costs per session that the airport network operator might incur at different usage levels:

On the revenue side of the equation, the wholesale network operator will collect money from the retail service providers whose end customers use the networks. Additional money will also be collected from ‘transient’ end users who pay by credit card. Given these estimates, the wholesale network operator can be profitable under the bad-case scenario, low usage and a high fixed payment to the airport, if wholesale roaming rates are $5 per session. Under the better-case scenarios with higher usage and lower property owner payments, the network operator is profitable even at roaming rates of $0.75 per session or less.

RETAIL SERVICE PROVIDER PERSPECTIVE
Now let’s look at the overall public Wi-Fi business opportunity from a retail service provider’s perspective. As a provider that maintains an ongoing direct relationship with the end user, a retail provider must also take into account costs such as customer acquisition and billing.

For this case study, we’ll look at the costs for a “virtual” retail provider, one who provides service entirely through roaming and does not own Wi-Fi networks. We’ll assume that the service provider already has a customer base and support infrastructure in place, and the public Wi-Fi service will be marketed as an extension to existing customers.

The main costs here include customer acquisition, customer support, billing, roaming partner acquisition, supporting equipment and wholesale roaming fees. The table below summarizes some cost ranges for these items that we might consider.

On the revenue side, the retail service provider may offer any number of end user pricing plans. Let’s look at an ‘average’ conservative scenario using the following values for key variables:

·        Customer take-up – 200,000 in year 1, growing by 50,000/year

·        End user pricing – $20/month (unlimited usage)

·        Average usage – 36 sessions/year/customer

·        Customer acquisition – $200/subscriber

·        Customer support – $30/new subscriber/year + $5/existing subscriber/year

·        Billing – $1,000,000 in year 1

·        Partner acquisition – $200,000/year

·        Equipment – $0.50/session

·        Weighted average wholesale roaming fees – $3/session

Key results of the model are illustrated in the figure below:

Given our fairly static set of assumptions, we get a predictably simple result. But it is interesting to note that net income turns positive in year two, with an initial investment of under $30 million (driven primarily by the $40 million in year one customer acquisition). The overall business has a net present value of several hundred million dollars.

We can vary these assumptions and add more complexity by accounting for price erosion, changes in customer usage patterns, volume-based roaming rates, etc. But under a broad set of conservative assumptions, we get similar outcomes in terms of rapid payback, low upfront investment and attractive ROI.

THE PROMISE OF PARTNERSHIPS
It is only by viewing the two models together that one truly understands the synergistic nature of public Wi-Fi networks. In order to drive volume and thus bring costs down, wholesale providers must turn to retail providers to obtain more customers. At the same time, retail providers, even those that choose to build out some of their own hotspots, can add to their bottom line by extending coverage through partnerships with Wi-Fi wholesale providers.

Only this roaming/partnership-centric approach -- where economies of scale are reached and end users are provided with broad coverage -- enables profitability for all players. And most importantly, profitability can be obtained even looking conservatively at only the ‘business traveler’ market opportunity. The potential to expand usage and deliver vastly greater profits by capitalizing on other opportunities once the infrastructure is in place -- a potential that is magnified by the continued incorporation of Wi-Fi into a wide range of devices -- is a topic for another day.

Kevin Jackson is co-founder and vice president of marketing at Tatara Systems, a developer of service delivery platforms that enable both retail and wholesale service providers to profitably support roaming relationships without sacrificing security, control or advanced capabilities. Visit Tatara’s Web site for more information.