Tag Archives: google fiber

Competition, and something more, drives Comcast upgrade in Huntsville

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

Demand.

Chalk up another win for broadband competition. Comcast announced that it was expanding its next generation – DOCSIS 3.1 – cable modem footprint to Huntsville, Alabama, and would be offering gigabit-level service to at least some customers. Details on service locations, roll out schedule and prices were lacking, though.

What clearly isn’t lacking is a competitive threat. Huntsville’s publicly owned electric utility is in the process of building a fiber to the home network that will be operated by Google Fiber and offer gigabit service at about half the price that Comcast charges in the four cities where it’s already offering it. Those cities include Nashville and Atlanta, where Google Fiber is also deploying fiber to at least some neighborhoods, Chicago, where Google-affiliate Webpass is present, and Detroit, which has neither.

Comcast similarly responded to plans in Santa Cruz to build a municipally-backed FTTH system by upgrading its plant.

AT&T previously announced that it would be offering gigabit service in Huntsville. It, too, has reliably followed Google Fiber’s lead as it prioritises the capital investments it makes in service and infrastructure improvements.

Although Comcast and AT&T are certainly playing defence and trying to prevent competitors from gaining a foothold, there’s also something like a virtuous circle effect going on. Google is – or, at least, was – identifying communities that were favorably disposed towards ultra-fast Internet service and then pumping up enthusiasm even further. For example, according to a story by Lee Roop on Al.com, Google reps spoke at a recent meeting of Huntsville entrepreneurs. One talked up the potential for small businesses and “another Google representative said homeowners can expect a $5,000 increase in their homes’ value if they add fiber optic cable”.

The more enthusiasm and awareness, the greater the market potential for high end broadband service. Competition feeds demand which draws even more competition. That’s how Huntsville is staying on the right side of the digital divide.

Welcome to Webpass City, California

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

We won. Why shouldn’t you believe it?

Google finally ‘fessed up to ditching its fiber construction business. In a blog post worthy of Baghdad Bob, the (now) former head of Google Fiber and related businesses – Craig Barratt – promised to “stay ahead of the curve — pushing the boundaries of technology, business, and policy — to remain a leader in delivering superfast Internet“. As he also announced his resignation.

By the time you read this, Google Fiber’s website might have changed, I’m sorry, pivoted again, but as it stands Californian cities are either transitioning from potential fiber city limbo to incumbent monopoly hell, or have been blessed as “Webpass cities”.

Which amounts to the same thing.

As the dust settled on Barratt’s mea culpa mission accomplished declaration, the Silicon Valley communities of San Jose, Santa Clara, Sunnyvale and Palo Alto were left in the potential category, where Google is “going to pause our operations and offices”.

Translation: we’re done with you, but we’re not going to actually say it because we’d lose whatever leverage remains over AT&T, Comcast and the rest.

Google is also “ever grateful to these cities for their ongoing partnership and patience, and we’re confident we’ll have an opportunity to resume our partnership discussions once we’ve advanced our technologies and solutions”.

Translation: so long and thanks for all the fish.

But San Francisco, where it had previously said it was in the fiber business, San Diego where it said it might some day be, and Oakland, Berkeley and Emeryville where it’s never previously shown any interest, Google is now proclaiming as “Webpass cities”.

Translation: we bought Webpass, we’re not sure why anymore, but we own those customers there, so we’re going to claim them and hope no one notices the difference.

The possible, but not likely, exception is Irvine, which Google still lists as “upcoming”. Given the dismantling of its fiber construction operations, that’s a distinction without a difference: if there isn’t fiber there already, there never will be. Not from Google, at least, but almost certainly not from any competitive private sector player either.

Salt Lake City may be debut of Google Fiber 2.0

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

Critical mass.

Google has launched what might be its last fiber project. Or maybe it’s the first deployment of Google Fiber 2.0. Residents and small businesses in the densely populated central area of Salt Lake City can now sign up for service, if they are in reach of the fiber plant that’s been installed.

As it typical, Google is hazy on the details of exactly where service is available, or what future expansion plans might be. However, one clue is the emphasis that Google is placing on apartments, condos and office buildings. According to a story in the Salt Lake City Tribune, apartment complexes had a head start on construction

In terms of residential customers, [Scott Tenny, head of Google Fiber business operations in Utah] said, single-family homes will require a visit from an on-site installer, while many apartment dwellers will be able to switch on service more simply.

Google Fiber has taken advantage of a steep increase in apartment building in recent months, Tenney said, and wired a large number of new residential complexes during construction for easier access once tenants move in.

At this point, Google’ Salt Lake City deployment has more in common with its limited plans for San Francisco, where it is specifically targeting multiple dwelling units within reach of existing fiber that it can lease from other carriers, than it does with full city builds in, for example, Kansas City or Austin. Or the full city builds in Silicon Valley and Portland that were put in a deep freeze while Google evaluates fixed wireless service via magic radios.

A mix of fiber to the premise builds in areas that are sufficiently densely packed to make a business case and wireless as far as it will go elsewhere is a rational strategy for maintaining hopes of an acceptable return on investment somewhere down the road. It’s a strategy that AT&T embraced a few years ago, when it decided to focus wireline investment on high potential areas, and then followed it up with an aggressive and ongoing push to move less profitable wireline subs to wireless-based service.

It’s a sharp departure from the vision of a fully fibered future that Google dangles in front of communities it’s courting. The gigabuck realities of fiber builds seem to have changed that vision.

Silicon Valley drops off of Google Fiber roadmap

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

Google has scrapped seemingly imminent plans to start laying fiber in Silicon Valley. According to a story yesterday in the San Jose Mercury News, Google has told officials in Palo Alto and Mountain View that the project is on hold, and the group that was to do the work has been disbanded…

The company was set to begin digging in San Jose last month, but nearly 100 employees hired to install Google Fiber were pulled into an office and told the project was being delayed, according to workers. They were offered a transfer to San Diego to work on an unrelated project.

The reasons given to the Mercury News are nearly identical to the statement Google made when it similarly pulled out of Portland

Google Fiber spokeswoman Veronica Navarrete said company officials will continue talks with San Jose, Palo Alto, Santa Clara, Sunnyvale and its hometown of Mountain View about providing internet service, but added that it “takes time” to deploy “the latest technologies in alignment with our product road map, while understanding local considerations and challenges.”

The latest technologies Google is considering include, maybe exclusively, wireless, the fruits of its acquisition of Webpass. On parent company Alphabet’s second quarter 2016 earnings call last month, chief financial officer Ruth Porat confirmed that Webpass’ wireless is now part of the mix at Google Fiber, and they are “continuing to push the frontier” with “different execution paths”.

Porat also said that total capital investment in what Alphabet calls its “Other Bets” was $280 million in the second quarter, “primarily reflecting ongoing investment in our Fiber business”. With the tab for building out a full fiber network in Silicon Valley looking like ten times that figure or more, and no end to operating losses in sight, Google’s cold feet come as no surprise.

More delays for Google Fiber hopefuls

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

Google is putting the brakes on its fiber builds. That seems to be the word out of Portland. According to a story in the Oregonian, contractors involved in the project – or at least who think they’re involved – say that construction won’t begin for several months, if ever. Google Fiber hasn’t actually said that Portland is one of its chosen few markets, but the general expectation was that an announcement to that effect would come in the fall.

The explanation a Google spokesman gave to the Oregonian indicates that the company is reevaluating its technology options…

“We’re continuing to explore the possibility of bringing Google Fiber to Portland and other potential cities,” Google wrote. “This means deploying the latest technologies in alignment with our product roadmap, while understanding local requirements and challenges, which takes time.”

In the context of Google’s recent purchase of Webpass, which does most of its business wirelessly, the term “latest technologies” doesn’t point to, say, slimmer fiber cables. More likely, it indicates a cost-benefit analysis is underway. The capital cost of installing fiber in major metro areas is huge, even by Google standards – the Oregonian puts a $300 million price tag on a Portland build.

Here in California, the tab would run even higher. The estimate for San Jose alone was in the gigabuck range and you can multiply that a few times to cover the rest of the Silicon Valley cities Google has been trawling. San Francisco is now listed as an upcoming Google Fiber city – in contrast to San Jose’s and Portland’s potential status – but that also points to a shifting business model. Google plans to lease other people’s fiber to reach multiple dwelling units there, and the Webpass acquisition complements that strategy.

Google Fiber finds a balancing point between home and business FTTP

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

Google Fiber is rolling out service plans for small businesses, with prices ranging from $70 a month for symmetrical 100 Mbps service to $250 a month for a symmetrical gigabit, all with no data caps. The price for a gig is considerably more than Google’s standard $70 a month residential rate, but it also allows for more bandwidth-intensive uses. Up to a point.

For example, the acceptable use policy for Google’s residential service clearly prohibits running an online business via the connection…

You agree not to use or allow third parties to use the Services provided to you for any of the following purposes…

To operate servers for commercial purposes. However, personal, non-commercial use of servers that comply with this AUP is acceptable, including using virtual private networks (VPN) to access services in your home and using hardware or applications that include server capabilities for uses like multi-player gaming, video-conferencing, and home security.

On the other hand, the small business acceptable use policy says nothing about operating a server for commercial purposes, except you’re not allowed…

To create substitute or related services through the use of or access to the Services (for example, to use the Services to provide web hosting services to third parties).

Nor can you resell Internet service, including offering it to tenants or hotel guests, except in common areas of your business. Google clearly intends to stay on the retail side of the business and maintain a direct relationship – include a direct, monthly bill – with end users.

A business that supports a couple dozen employees will consume more bandwidth than a typical household that can occasionally burst at higher speeds. Running a public facing server will cost a little more – the benchmark price for the necessary static IP address is $20 a month for one, $30 for five. But assuming Google has enough backhaul bandwidth to reliably deliver symmetrical 100 Mbps speeds to a business, the trade off between the low end small business package and gigabit residential service – both priced at $70 – is a fair one.

San Jose cuts a fiber deal but Google won’t say yes yet

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

San Jose is all for it, but Google Fiber remains coy about whether it’s going to build a fiber to the home system there, or elsewhere in the south Bay Area. On Tuesday, the San Jose city council voted unanimously to approve a construction plan and five fiber hut site leases on city land, for a prospective Google Fiber buildout.

Jenna Wandres, the Google representative at the meeting, said that they plan to build out to virtually the city, with the only possible exceptions physically hard to reach locations in the hills. She said they plan to build 2,300 miles of fiber plant – more or less the total street mileage in San Jose – regardless of household income levels in neighborhoods. That’s not a ironclad guarantee, though. It leaves the door open a crack for Google to prioritise builds on the basis of pre-sales or other consumer demand metrics, as it has done in other markets.

Pricing is not set yet, but it’s likely to follow the pattern set in the southeastern U.S., according to Wandres. There, Google charges $130 a month for a gigabit plus TV service, $70 for a gig alone and $50 for 100 Mbps service. Wandres said that there will be a fourth price plan, at least for low income neighborhoods, but it hasn’t been determined yet.

The construction plan calls for 40% of the network to be installed on utility poles, and the remaining 60% will go underground, with both traditional trenching and microtrenching techniques used.

The City of San Jose is getting about $5,000 a year in lease payments for each of the fiber hut sites. That’s more than twice the rate Google is paying in San Antonio, for example.

The City of Mountain View approved similar plans earlier this month. Google has not yet said, though, whether it will actually implement any of the approved plans. Wandres said that if they do decide to move ahead in the south Bay Area, it would be Google’s first “organic build” in California. San Francisco is the first California city with a Google Fiber project, but it’s mostly using existing fiber lines and involves little or no new construction.

No word on when a decision is expected.

Google Fiber tries cheaper service tier in Kansas City

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

Google Fiber is pulling its 5 Mbps data plan off the market in Kansas City. It’s often mischaracterised as free, but it wasn’t quite that. The deal was that people living in a soon-to-be blessed fiberhood could pay a $300 installation fee and get 5 Mbps service for seven years. Speculation is that Google Fiber is trying to pump up revenue by steering low end subs to a $50 a month, 100 Mbps plan with free installation.

That might be true enough, but I think there are a couple more important factors at play. First, indications are that the $300 deal wasn’t very popular. A survey done by Bernstein Research in 2013 estimated that the 5 Mbps tier was accounting for maybe 10% to 15% of Google’s subscriber base. That makes sense: people who are happy with 5 Mbps are not prime candidates for fiber to the home service. Some people will find the deal attractive, but it won’t be a big draw.

The second factor is multiple dwelling units – MDUs. Google has been battling it out with landlords, trying to get them to pay the $300 installation fee for each apartment in a building, on an all or nothing basis. Landlords would write a big, upfront check, residents would get the slow free tier, and if they upgraded to gigabit service, Google would rebate the installation fee a little bit at a time over the course of a year. I never understood Google’s thinking on that. Building owners would be likelier to want Google to pay a big upfront fee, and then share revenue forever. It’s how they think.

If Google provisions the 100 Mbps tier the same way it does for gigabit service, it’s a screaming good deal for the price sensitive side of the market. The selling proposition matches cable speed promises and then actually delivers, all at a DSL price.

Austin and KC owe Californians a thank you card for paying their Internet bill

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

The gift that keeps on giving.

The fiber to the premise analysis done by the City and County of San Francisco summed up the likely competitive response to a municipal build in two words: Google Fiber. Incumbents in the markets Google targeted responded with upgrades and lower prices…

The incumbent providers’ responses to Google Fiber’s expansion in other cities may foreshadow their responses to a municipal network in San Francisco. After Google Fiber came to Kansas City, incumbent providers Comcast and Time Warner upgraded their networks to double residential speeds, which lowered the dollar per megabit cost of bandwidth for their customers. Industry experts interviewed by the Budget and Legislative Analyst estimate that these upgrades cost very little for incumbent providers and occurred because of the competitive threat Google Fiber posed. Similarly, AT&T deployed its high speed Gigapower network in Austin shortly after Google Fiber announced its intention to build in that City. AT&T is charging $50 less per month for Gigapower in Austin, where it competes with Google Fiber, than in other cities where it does not. Similarly, Comcast upgraded residential speeds at no additional cost to customers after the City of Santa Cruz announced its intention to form a public-private partnership to deploy a FTTP network.

AT&T isn’t lowering prices where its monopoly holds fast. Comcast and Time Warner aren’t raising speeds at no additional cost where their monopoly (or duopoly) status is secure. The difference between what AT&T, Comcast and the rest charge in a competitive market and what they charge everybody else is either pure profit, or a mix of profit and cross-subsidies into competitive areas.

That’s why AT&T, supported by the cable lobby, is fighting so hard to lock potential competitors out of taxpayer-funded broadband subsidies in California, and keep the money for itself. If they win, you’ll pay twice: once for the subsidy, and then forever for monopoly-priced service.

Hints of respectability for Google Fiber’s sub count

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

Some tantalising numbers have been published that could be interpreted as bad news for Google Fiber’s subscriber count. Or it might foretell market success. It depends on how you look at it.

MoffetNathanson Research took a bearish look at the latest filings with the U.S. copyright office, which collects pay TV data, and found that Google has 53,000 video subscribers in the three markets where it’s doing business – Kansas City, Provo and Austin – which indicates it grew by only 23,000 subs in 2015.

That’s video subs. Not broadband customers. Nor does it say much about how extensively Google has built out its fiber network in those markets. According to a story in re/code, Google Fiber has a reasonably good sized share of homes in the two Kansas Cities…

Moffett reports that Google has more than 7,000 video subscribers via its Google Fiber service in Kansas City, Kansas — about 13 percent of the market there — and more than 20,000 subscribers in Kansas City, Missouri — about 10 percent of that market.

Combine those numbers and you get about 11% video penetration of the two KCs (that’s within city limits – suburbs aren’t included). To me, that points towards a very healthy penetration of Google’s addressable universe, the homes it actually reaches with fiber.

Let’s play with the numbers a bit. If you make the wild ass guess that Google has built out half of the two cities and that half of its broadband customers opt for the video package too, then that 11% becomes a 44% share of its addressable broadband universe, which is a dominant share. If you assume an 80% broadband adoption rate – still a guess, but not so wild – that leaves the incumbent telephone and cable companies with 36% of the market, or 18% each assuming an even split. Ouch.

If you’re less wild and assume that Google has built out to three-quarters of the homes in the two cities and sells video to three-quarters of its broadband subs, you get a 19% take rate which is respectable for the third major ISP to enter the market. My guess is that the correct figure is somewhere between those two scenarios. Not dominating, but punching equally with cable and telcos: a success for Google Fiber.