Now that Google Fiber is to be built in three U.S. cities, and given AT&T’s announced intention to counter with a 1-Gbps network of its own in Austin, Texas, new attention will be paid to precisely what has to be tweaked in an ISP’s business plan to make such networks possible on a wider scale.
Other ISPs are taking a look at the economics as well. Wicked Broadband in Lawrence, Kan., for example, is among the latest to announce it will build a 1-Gbps network.
Work will have to be done on both the revenue and cost fronts, since the retail pricing set by Google Fiber ($70 a month for 1 Gbps, free 5 Mbps service) disrupts current pricing levels. Where available, 1-Gbps connections have sold for about $300 a month (Utopia in Utah) (EPB Fiber in Tennessee) (Sonic.net in northern California).
With a $70 retail price for 1 Gbps, all lower speed services likewise will have to be reevaluated.
That is going to compress profit margins for any ISP that actually does boost speeds, even if speeds are not increased all the way to 1 Gbps immediately, since the “1 Gbps for $70” pricing umbrella almost inevitably will require a revision of all lower-speed prices as well, as that price point is lower than what ISPs now set for 50 Mbps services.
So that’s one level of new ISP challenges. The related challenge is how to scale operations and marketing to match revenue that inevitably will have lower profit margins. In part, Google Fiber has started a trend, where neighborhoods get built depending on the actual level of demand from households in those neighborhoods.
That is likely to help with cash flow, but probably will not affect the long term requirement to build out a full network, passing every household, within a specified number of years, at least for any ISP offering video entertainment services requiring a video franchise.
The point is that marketing costs will have to scale in some way to match the lower profit margin that 1-Gbps access, sold for about $70 a month, will offer. The same logic will apply for operating and customer service costs as well.
And though costs to deploy any fiber to the home network have dropped significantly over the last couple of decades, cost reductions will be stubborn, as the costs of construction have not dropped, and those costs are the key drivers of total cost.
An ISP with huge scale, such as Verizon Communications or Comcast, will be able to leverage that scale. But most smaller ISPs will probably still find that after the cost of network construction, connecting a paying customer will cost between $600 and $1000.
ISPs who do not offer video entertainment services might well find the cost of connecting a high speed Internet access customer lower, possibly in the $450 range. Google Fiber itself uses a $300 retail connection fee, but that does not necessarily mean the “cost” is $300.
Among the big variables is the cost of customer premises equipment, not just the cost of drop cables, termination enclosures and hardware, plus installation labor.
ISPs large and small will have to begin reevaluating their basic assumptions about revenue and cost to provide high speed Internet access. Those assumptions now will have to account for likely increased network investment, different and lower profit margins for every tier of access service, and new requirements to adjust all marketing, overhead and operating costs to reflect the new business case.
And that will be the good news, in many cases. Though few ISPs will welcome the disrupted expectations about value for money, some service providers at least have a viable — albeit expensive — upgrade path.
For some ISPs, the gigabit access context will be very bad news, simply because the upgrade path does not exist. Satellite broadband and most fixed wireless providers might eventually find they simply cannot scale to match the new standard for access speeds and pricing.
Ironically, the shift in context to gigabit networks will have profoundly uncomfortable repercussions for many segments of the ISP business.