In the most significant announcement since SBC acquired the old AT&T and became “the new” AT&T, the telco giant announced it will spend $14B over the next three years to expand its wireline and wireless networks under its newly coined “Project Velocity” initiative. The company wants to move to an all IP network platform, which means they’ll be phasing out TDM transmission and the PSTN.
Surprising most analysts, AT&T said $6B of that $14B will be spent on wireline upgrades. In particular:
1. Residential Broadband via U-Verse, IP-DSLAM and (in some rural areas) LTE:
Traditional U-Verse (TV, high-speed Internet, VoIP) as well as U-Verse IP-DSLAM (high-speed Internet, VoIP, but NO TV service) will be available in many more areas with Internet access speeds of 75M b/sec for most customers, with many achieving speeds of up to 100 M b/sec (downstream).
Currently, 32% of AT&Ts customers are covered by (triple play) U-Verse, which will increase by one-third to 8.5M additional customers and 43% coverage by the end of 2015. U-Verse revenues are running at a $9B annual rate and increasing at a 38.6% annual rate. AT&T is making “customer retention improvements in all areas.” (Presumably to avoid losing U-Verse customers to triple play MSO services, e.g Comcast/Xfinity which runs commercials enticing U-Verse customers to come back to Comcast for better high speed Internet and TV service).
AT&Ts wired IP broadband network will expand to 75 percent of residential customer locations in AT&T’s 22-state wireline service area by year-end 2015. Not all of those potential customers will be able to get U-Verse TV service. Rather, they will be connected to IP DSLAMs to achieve higher speed Internet access. (This means AT&T will be jettisoning its ATM over ADSL network in favor of IP/Ethernet transport to/from customer premises to its initial point of presence where the DSLAM resides. Customers currently using ATM over ADSL will have to be retrofitted with new CPE to access U-Verse IP DSLAM or traditional U-Verse).
The higher Internet access speeds (over last mile copper) for U-Verse and IP DSLAM will be achieved by VDSL pair bonding, “small form electronics,” and VDSL vectoring.
Status Report: Currently about 9% of AT&T landline customers cannot get broadband, while 32% have the U-Verse triple play available to them, 32% have U-Verse IP DSLAM available and 27% are served via legacy broadband (i.e. non-IP DSL).
By the end of 2015, AT&T said 99% of customers in its 22 state service area will have broadband available to them via either via wireline or LTE option. The breakdown is as follows: 43% of customers will have access to the U-verse triple play, 32% will have access to the U-verse IP DSLAM and the remaining 25% will need to rely on LTE, which will be available to 99% of AT&T’s customer base.
Other key points related to residential broadband are as follows:
- Much higher speed wireline Internet, via either U-Verse triple play or U-Verse IP DSLAM (double play) will be available to 57M AT&T customer locations by 2015. The IP DSLAM double play offering of broadband Internet and VoIP will be available to 24 million customer locations by year-end 2013. Those customers will be offered a triple play bundle based on IP DSLAM and satellite video service from Dish network.
- Customers in rural or remote locations (presumably the 25% who won’t get wireline broadband access) will be able to get LTE wireless access, as AT&T plans to extend its 4G LTE build out to cover 300M POPs by end of 2014.
- High speed IP connectivity will be available to 99% of wireline service?area customers (via either U-Verse, IP-DSLAM or LTE) by 2015.
Summing up, AT&T’s firmly believes that:
- Wireline IP broadband is structurally attractive in dense population areas
- IP broadband is the most important product in the triple or quad-play bundle
- AT&T IP broadband will meet customers’ growing speed requirements
- Significant synergies exist between wireless and wireline assets
2. Fiber to the Building deployment:
AT&T will light fiber to reach 1 million additional business customer locations, covering 50 percent of multi-tenant office buildings in AT&T’s wireline service area by year-end 2015. 50% of those multi-tenant office buildings in AT&Ts wireline service area will be fiber connected. (That’s up from about 15% nationwide today).
3. Strategic Business services:
IP VPN, Carrier Ethernet, (Web server) hosting and vaious managed business services generated $6.4B in revenues last year and is growing at 14.5% annually. Wireline data and managed IT services for enterprise customers are growing at a rate in excess of 6%.
Cloud computing and security are seen as the next big growth opportunities as AT&T transitions to managed services for its enterprise customers. In particular, AT&T plans to partner with cloud service providers as well as providing cloud services over their own managed IP network that leverages performance, reliability and security. During the Analyst Day webcast, AT&T said, “Virtualization and mobilization are driving the need for a ubiquitous, dense wireline footprint solutions that bundle cloud with connectivity (AKA Cloud Networking), symmetrical bandwidth, and security through active network management.”
Century Link/Savvis and Verizon/Terremark are recognized cloud leaders each having very solid cloud computing with managed IP VPNs for delivery of cloud services. They will now have much more competition from AT&T in the cloud space.
Additional information on U-Verse:
As indicated in the graph below, AT&Ts broadband market share is growing in areas where U-Verse is available to residential customers.
U-Verse has delivered 5 years of top line growth for AT&T:
- $9.5B revenues, which are growing 38% Year over Year
- 7.1M IP broadband subscribers, with 2.5M added in last 12 months
- 4.3M IPTV subscribers, 760K gained in last 12 months
- 18% U-verse video penetration; 23% U-verse broadband penetration
- ~$170 ARPU for U-verse triple-play service bundle
At a Wells Fargo investment conference on November 8th, AT&Ts VP & CFO John Stephens said that AT&T evaluated commercial buildings with six tenants or more to determine whether they should get fiber connected. Considerations included: distance from AT&Ts central office (CO), cost efficiency and build-out cost.
A huge side benefit for AT&T is that once the fiber to the building is installed and AT&T owns the right of way, the company will install Distributed Antenna Systems (DAS) along the fiber route to provide increased 3G/LTE wireless coverage. The DAS’s would use fiber backhaul to AT&Ts CO. Mr. Stephens hinted that DAS’s (deployed along the fiber-to-the-building route) might also be used for broadband wireless offload, but did not disclose any details how that might work or be configured.
[Infonetics analyst Stéphane Téral recently said in an email, “The majority of operators are still using distributed antennas (DAS) in their mobile networks for coverage, and despite all the talk about using small cells to boost capacity in large venues, operators we interviewed believe DAS will remain a fundamental tool for malls, airports, stadiums and the like.”]
Mr. Stephens was both enthusiastic and confident during his presentation. He said, “AT&T is investing in tried and true things we know. We are moving away from PSTN and torward an all IP network platform for delivery of all telecom services including voice.”
Closing Comment:
During its November 7th Analyst Day webcast, AT&T CEO Randall Stephenson echoed Mr. Stephens confidence, “These are things we’ve done before – logical extensions of proven technologies and already successful businesses. We are very confident in our ability to execute this plan.”
Reference:
http://www.att.com/gen/press-room?pid=23506&cdvn=news&newsarticleid=35661
11 replies on “AT&T to Expand U-Verse & IP-DSLAM; Bring Fiber to Commercial Buildings & Cover 99% of US with LTE!”
Thanks Alan for posting your insight into this well publicized and, as you suggest, significant announcement.
This is a story that has many interesting angles that could be explored in greater depth. For instance in the rural area, it looks like they will be divesting those areas where they can’t serve up broadband over wireline, based on slide 33 of the analyst’s presentation:
http://viodi.com/?attachment_id=7957
That same slide also suggests that the wireline and wireless business would be separated, which sort of sounds “back to the future” (e.g. s you point out, AT&T wireless was acquired by SBC some years back).
Back to the rural areas, what will happen to the exchanges and customers that they divest Who will bid on those properties and at what price given that AT&T doesn’t think it makes sense to upgrade? They did reject the CAF money to upgrade rural properties, so that’s one benchmark for the value of some of these exchanges and outside plant.
http://viodi.com/2012/07/26/top-us-telcos-reject-broadband-connect-america-funding/
Much like Verizon did with their rural properties, it looks like AT&T is banking on serving the hard to reach areas with LTE (potentially competing with the people who acquire their rural wireline areas).
It is also an interesting statement that the upgrades will cover 75% of their customer locations across 35% of the geography? Looking at the maps, that must mean the 35% of the geography of the 22 states. If the upgrades are covering 75% of customer locations, does that mean the other 25% will be ones that are potentially going to be divested?
Lots of questions that I am sure will be inspire more articles over time.
Ken, Thanks for your comment and keen analysis of AT&Ts wireline coverage plans.e.g. “the upgrades will cover 75% of their customer locations across 35% of the geography in AT&Ts 22 state area.”
I have no clue as to whether AT&T customers that are NOT scheduled to get wire-line broadband will see their existing copper based telecom services go dark or divested to a local telco. On Tuesday of this week, I talked to an AT&T splicing technician who was reconditioning copper lines on telephone poles (“pro-active maintenance”) in Blue Lake Springs (BLS), CA (altitude 4500 feet and near the town of Arnold, CA).
He said that neither U-Verse or IP DSLAM were coming to the BLS area anytime soon and that AT&T might sell their copper plant there (inherited from pre-divestiture AT&T, then PacBell & SBC) to a 3rd party. He gave no hint that phone service would be turned off if AT&T could not sell their local copper plant. He also didn’t know when or if AT&T wireless would have a presence in the BLS area, which is currently served by only one wireless carrier- VZW (which offers voice and 3G, but not LTE service). If not, then BLS residential and business customers would not get either wire-line or wireless broadband from AT&T in any way, shape or form.
The principal broadband Internet provider in the BLS area is Comcast/ Xfinity. AT&T offers ATM over DSL (up to about 2M b/sec downstream) in lower elevation sections of BLS. To entice BLS residents to buy iPhones, AT&T gives them a free femtocell, which wirelessly connects their iPhone to the broadband wireline network (mostly Comcast/Xfinity cable modem) for all their Internet traffic.
The Wall Street Journal reported that AT&T is working on a state-by-state basis to overturn regulations that require them to offer phone service wherever the carriers own wireline networks. The article also suggested that investment would go to those states that are more flexible in changing regulations; this is a similar strategy employed by AT&T in its push several years ago for statewide video franchise laws.
Geoff Burke of Calix has a good post on AT&T’s investment plans and the implication for rural areas and rural carriers. He makes an important statement,
“…..now is the time for all service providers adjacent to or serving these communities to begin thinking about this option and their strategy.”
Click here for the link to Burke’s post, which has a number of good points about AT&T’s plan as it relates to the National Broadband plan and CAF
http://community.calix.com/t5/Calix-Community-Blog/AT-amp-T-Shows-Its-Hand-Now-What/ba-p/19696
“Those customers will be offered a triple play bundle based on IP DSLAM and satellite video service from Dish network.”
One also has to wonder if AT&T will finally make a play for DISHTV; almost 8 years later.
http://www.viodi.com/newsletter/050200/#dish
I doubt that AT&T would make a bid for DISHTV, so many years after they started reselling Dish’s satellite TV service. For many years, that’s the only TV service you could buy from AT&T. In fact, Santa Clara, CA – the center of Silicon Valley- only got extensive U-Verse TV & high speed Internet coverage this summer! Before that the only double/triple play available in Santa Clara was from Comcast/Xfinity!
The plan of AT&T to expand U-verse and deploy fiber deeper in the network is very sensible. This is the most economical way to greatly expand broadband speeds, and is similar to plans of large European providers. Based on their stated targets to achieve speeds in the region of 75-100 Mbps, it is clear that AT&T plans to deploy vectored VDSL2 technology. The following video gives some information on vectored VDSL2:
http://t.co/kDnBNI6n
Thanks George for the comment and link. You are right that vectored VDSL2 technology will be central to what they are doing. I had a hand in producing a video that showed the difference with vectoring on and off, both in a lab and real-world condition.
http://portal.calix.com/solutions/dsl.html
Clearly, this will help with the shorter drops, particularly as other disturbers in the binder group are removed and all the signals are transitioned to VDSL2.
AT&T coverage issues have been generally known. As one reviews the technology components within this plan one has to wonder why was it not put in effect 2+ years ago! Investment issues and economics have not changed significantly.
Still, I hope they execute this time.
It appears AT&T will not build-out U-Verse or IP DSLAM to rural areas, but instead give many (but not all) rural users access to 3G/4G-LTE wireless broadband. That will require many who only use desktop or notebook PCs to buy smart phones or tablets. Furthermore, 3G/4G wireless access may be much more expensive than unmetered (“all you can eat”) wireline broadband. Data plans are more expensive and there are data caps with overage charges for current AT&T mobile subscribers.
One huge surprise about AT&Ts wireless expansion is it told the FCC & US regulators that it couldn’t build out its LTE newtork to rural areas WITHOUT acquiring the spectrum and cell towers owned by T-Mobile. Many pundits are now saying that AT&T lied about that. Here’s an example:
When AT&T Said It Needed T-Mobile For LTE, It Wasn’t Being Completely Honest
http://www.tmonews.com/2012/11/so-when-att-said-it-needed-t-mobile-for-lte-it-wasnt-being-completely-honest/
An IEEE ComSocSCV Discussion group member sent me an anonymous comment relating to your last paragraph. Subject line read: “AT&T caught in a brazen 4G lie”
http://money.cnn.com/2012/11/07/technology/mobile/att-4g/index.html
He then sent me another related comment with Subject line: “Lies, lies, lies…”
Reference was made to this article:
AT&T Admits That The Whole ‘Spectrum Crunch’ Argument It Made For Why It Needed T-Mobile Wasn’t True
Here’s an excerpt: “AT&T is now saying publicly that there is no spectrum crunch. It has more than enough.”
Speaking to analysts, AT&T’s chief strategy officer John Stankey yesterday acknowledged the company is now well-positioned on the spectrum front — even before the company starts moving on their new plan to use WCS spectrum for LTE deployment.
“Even under ideal circumstances, getting new spectrum on the market in the next five to seven years is aggressive,” Stankey said. “But what we do know is that AT&T is well-positioned now…These deals give us confidence that we can meet our LTE objectives for next two years and they will allow us to deliver competitive performance.”
Full article is at: http://www.techdirt.com/blog/wireless/articles/20121109/07434820984/att-admits-that-whole-spectrum-crunch-argument-it-made-why-it-needed-t-mobile-wasnt-true.shtml
Thanks for this very informative article with lots of discussion.