Telecosm - Ike Elliott

Worthless Guidance? Neutral Tandem's Sandbagging Ways

Neutral Tandem, the wholesale provider of tandem switch interconnection services, exceeded estimates and raised guidance when it announced earnings earlier this month. The company announced revenue of $28.6M in the 2nd quarter, continuing a steady stream of quarter-over-quarter revenue increases.  However, like many public companies, Neutral Tandem's guidance doesn't really tell us much. 

The company raised its prior guidance for the full year 2008 to somewhere between $117M and $121M in revenue, higher than prior guidance of $112M to $116M.  To put this into context, let's look at the company's revenue results by quarter for the past eight quarters:

Neutral Tandem	3Q06	4Q06	1Q07	2Q07	3Q07	4Q07	1Q08	2Q08
Revenue (millions)	$  13.7	$ 15.0	$ 17.6	$ 20.5	$ 22.6	$ 24.8	$ 26.2	$ 28.6
QoQ Change		$   1.3	$   2.6	$   2.9	$   2.1	$   2.2	$   1.4	$   2.4

The company has been adding new revenue in the range of $1.3M to $2.9M per quarter over the past two years, so the new guidance of $117M to $121M in 2008 revenue looks to be in the bag.  The company already had $54.8M in revenue in the first two quarters, so to make the low end of guidance, they need to add an average of $1.5M in new revenue for each of the next two quarters.  To reach the high end of guidance, they need to add an average of $3M in new revenue for each of the next two quarters. 

This range of performance on revenue growth of $1.5M to $3M in new revenue per quarter covers almost the entire range of quarterly revenue growth throughout the company's history.  The company is basically telling us "We won't perform any worse than the worst quarters in our history, and we probably wont' perform any better than the best quarters in our history, but between those two extremes, we can't tell you where we will land." 

So, why raise guidance at all, then?  Well, it was pretty clear by the end of the 2nd quarter that the company was going to blow away their previous revenue guidance, so the company had to revise guidance.  To say nothing would cause the street to assume that the old guidance still stood and that the company was forecasting a revenue slowdown because they didn't raise guidance. 

But the company was sandbagging with that old forecast, anyway, and the street knew it.  The table below shows that at the low end of the previous guidance, $112M for 2008, that at the end of the 2nd quarter all the company needed to do to make guidance was repeat it's 2nd quarter revenue of $28.6M for two more quarters, for zero revenue growth in the second half.  That certainly wouldn't be enough to get anybody excited.

All figures in millions
Forecast 2008 Revenue	$  112	$  116	$  117	$  121
Remaining Revenue	$ 57.2	$ 61.2	$ 62.2	$ 66.2
Average Quarterly Revenue Needed In 2nd Half To Achieve Forecast	$ 28.6	$ 30.6	$ 31.1	$ 33.1
Actual 2nd quarter Revenue	$ 28.6	$ 28.6	$ 28.6	$ 28.6

Neutral Tandem isn't alone in their sandbagging ways, so I don't really mean to pick on them alone.  Their 2nd quarter earnings were actually pretty good.  I'm just pointing out that most public companies bracket their guidance with ranges that they are absolutely certain they will meet, at least at the time they make the forecast.  Yes, now and then we do actually find surprises in guidance, but to be honest, most of the time I find guidance to be of limited value. 

      

Podcast - August 26, 2008 - New Global Telecom. Guest: John Guillaume

How does the largest wholesale hosted IP PBX service provider think about the market?  Today's podcast interview is with John Guillaume, Senior Vice President of Product and Marketing for New Global Telecom.  Listen in as John gives us some insight into NGT, their recent work with Microsoft, and the their perspective on opportunities in wholesale VoIP services. 

  (About 35 minutes)

Download guillaume_ngt_interview.mp3

VoIP on a Plane: Cool, Disruptive, or Dangerous?

The news that Andy Abramson successfully used Phweet to skirt the rules prohibiting the use of VoIP on airplanes has certainly created a stir.  Reaction of commenters has been wide-ranging:

• Some thought it was a cool demonstration of how hard it is to control what happens on an IP network.
• Some reacted with revulsion at the thought they would have to sit next to loud talkers for the duration of the flight.  Some called for the creation of "talking" and "no talking" seating sections on the plane.
• Some even thought that the use of VoIP from a plane would make it easier for terrorists to coordinate plane-based attacks. 

I'm not sure about that last one...Internet chat should certainly work about as well as VoIP for a terrorist's purposes. 

The more interesting question is whether it is rude to talk on a phone (or PC or PDA) while sitting next to somebody on an airplane.  I don't know.  Suppose you are in a middle seat and you strike up a conversation with the person to your right.  Is that rude to the person you are talking with?  It would be pretty easy for the person on your right to give clues to you that they would like to end the conversation, though. 

The bigger question is the impact to the person on your left.  Suppose you are happily chatting with the person on your right for an hour or more.  How does the person on your left feel about that?  Should you ask them or just assume that you should limit your conversation, assuming the person on your left would be annoyed with an extended conversation? 

These questions of airplane courtesy are important because they bear a similarity to the question of VoIP on planes.  My take:  you should get a pretty solid read on what your seatmates are willing to put up with before doing any extended VoIP calls. 

What do you think? 

VoIP is Still Disruptive

There is one thing about VoIP fanatics that you can count on:  show them a "rule" and they will find a way to break it.  That's what Andy Abramson and Joanna Stern did last night.

American Airlines is launching a wireless Internet access service called Gogo, provided by Aircell, on some of their flights, providing unlimited Internet access for the duration of the flight for about $13 per flight. Unlimited, that is, except for this little gotcha from their terms of use:

"No Voice Applications.   You will not use any type of voice application (including, without limitation, voice over Internet protocol) without written permission from Aircell."

Various motives have been ascribed to Aircell for this restriction, including a desire to keep people from bothering other passengers by engaging in an interactive conversation with somebody not on the plane, or, more likely, a desire to preserve ridiculous profit margins on pre-existing in-flight telephone services.  Whatever the motive, Joanna and Andy were determined to break the rule.

Joanna Stern, News Editor for Laptop Magazine, was on an Aircell-equipped American Airlines flight last night, and decided to try a VoIP call, inviting Andy Abramson to join her using SightSpeed.  Well, Aircell blocked that call but it didn't stop Andy.  He kept trying, and guessed rightly that Aircell wouldn't be blocking Adobe Flash.  Knowing that Phweet's VoIP application used Flash audio, he invited Joanna into a Phweet conversation and they happily had an extended VoIP conversation for most of the remainder of the flight. 

This reminds me of those locked iPhones from last year...it wasn't long before there were a dozen or so work-arounds for unlocking an iPhone so that it could be used with a non-AT&T network.  Where there is a will, there is a way.

Kudos to Andy and Joanna for proving that you can't keep a good VoIP down.

The Fate of Over-The-Top Business VoIP

This week I've been exploring over-the-top business models for residential (Vonage) and business (8x8) markets.  Over-the-top residential VoIP services are really hurting right now, so it made me wonder whether over-the-top business VoIP services would follow suit.  The answer: yes and no. 

As commenters on this blog have pointed out, Vonage is hurting for several reasons:

• Vonage offers only VoIP service, and generally can't compete with the triple-play of VoIP, Internet Access, and cable TV services offered by the cable companies.
• Also, the cable guys do piggy-back marketing, where a single mailer sells all three services in the triple-play, which can reduce the cost of sales for the cable companies.
• Vonage doesn't own the network so they can't control the quality of the network, and the cable guys do, so Vonage VoIP services are sometimes lower quality than competing cable services. 

Vonage has had a few other mis-steps (patent lawsuits, bungled IPO, etc.), but I think the three bullets above capture the biggest causes of their troubles.

How many of these troubles transfer to 8x8, who is betting on the over-the-top business VoIP market?  Well, for one thing, the video part of the triple play isn't of interest to many businesses.  Yes, some businesses could use cable TV services, but not as many as Comcast would like.  I paid a visit to the Cablelabs campus recently, and yes, they had a half-dozen TVs in their lobby, but consider the context.  Having a TV blaring in your lobby just isn't a requirement for many of your larger businesses.  Restaurants? yes.  Barber shops?  yes.  Law offices? not so much. 

So, cable's triple-play assault on the business market will appeal to some market segments, but the attaction will not be universal.  My guess is that the triple play will get the most traction on the small end of the small-to-medium business segment, as measured by site size. 

However, businesses will be very interested in the double-play of Internet access and phone service.  That bundle has been finding traction for a good while now, as demonstrated by Cbeyond.  (In fact, Cbeyond has created a triple play of their own, adding wireless phone service to the traditional double play bundle). 

8x8, on the other hand, doesn't offer the Internet access piece.  Nor does 8x8 offer wireless phone service, though they do have a way for you to save money on international wireless calls with an 8x8 plug-in on some phones. (The software notices that you dialed internationally, and places a call to an 8x8 VoIP gateway instead of calling the destination directly, siphoning the minutes off to a cheaper network).  So, 8x8 doesn't offer the complete bundle like Comcast or a Cbeyond, and as a result they are at a disadvantage, but not as deep a disadvantage as Vonage, since many business customers don't care about video. 

The second big disadvantage for 8x8 is that they don't own their own network and don't manage the quality of the voice packets on the network.  That will be a problem for customers with skinny or congested Internet access services.  The inconsistent audio quality will cause some customers to be dissatisfied with the service, which will contribute to customer churn rates.

Net-net, once 8x8's subscriber base comes to be dominated by business customers instead of residential customers, I expect their churn rate to be lower than Vonage's but higher than Cbeyond's.  Churn will be lower than Vonage because business customers churn less by nature, and because the video part of the triple play doesn't have universal appeal in the business segment.   Churn will be higher than Cbeyond because Cbeyond manages quality and offers a broader bundle, while 8x8 does not manage quality.  The long-term prospects for 8x8's business VoIP service depend on how low they can drive their churn rate.  The lower churn goes, the higher 8x8's subscriber base will be. 

VoIP Business Models and Churn: Cbeyond versus 8x8

Yesterday, I voiced my fears that the residential over-the-top primary line replacement VoIP business model is dying.  SunRocket died last year.  Vonage is struggling now.  What are the chances that the illness afflicting over-the-top residential players may spread to over-the-top business service providers? 

The chart below shows some of the business model choices of VoIP service providers:

When deciding to offer VoIP service, one of your first choices is around your network.  Are you going to build your own network?  For the cable providers, this was an easy choice.  They already owned their own network and just needed to extend its capabilities a bit to support VoIP service.  The problem for most companies, though, is that they don't already own a network, and networks are very expensive to build.  So most VoIP service providers choose to use somebody else's network.

When using somebody else's network, there are two choices:

• You can pay a network owner to use their network to connect to your customers.  This usually takes the form of buying network access from an ILEC or CLEC on a wholesale basis, then reselling that network access to your retail customer.  CBeyond and Unity Business Networks do this.
• You can assume that your customer already has public Internet access, and that somebody else is selling that to the customer, so you can just ride that existing Internet link for your VoIP application. This is called an over-the-top VoIP service, and Vonage and 8x8 do this. 

For the service provider (and the customer) there are trade-offs in this choice.  For one thing, riding somebody else's network, instead of reselling it, is free, so the service provider sheds the entire cost renting somebody else's network to get to the customer.  The savings is passed on to the customer, but the customer is left to buy their own Internet access a la carte, so that eats into the savings from the customer's perspective.  The big negative, though, is that when your VoIP service provider rides somebody else's network infrastructure, the quality of the voice service is unmanaged. 

What does it mean when I say that an over-the-top VoIP service is unmanaged?  It means that packets carrying audio for phone calls have no priority of packets associated with web browsing, email, and file transfers.  A burst of non-voice traffic can cause voice packets to be delayed or dropped.  If your Internet access has limited capacity, or if your higher-capacity Internet access is sometimes heavily utilized, then the quality of your phone calls will audibly suffer.  Some over-the-top service providers try to compensate for this with technology that can adapt to packet loss, such as adaptive codec technology, but even this technology can't make up for a heavily congested Internet link.

Since business customers are often more demanding than residential customers when it comes to voice quality, many business service providers avoid the over-the-top business model, choosing instead to rent somebody else's network.  By doing this, the service provider gains control of the Internet access, and can prioritize voice packets over other kinds of traffic, so that the quality of phone calls can be preserved. 

However, renting somebody else's network definitely increases costs for the service provider.  Renting a T1 in most U.S. cities of any size, even on a wholesale basis, can cost anywhere from $100 to $400 per month, depending on the city.  Longer distances can cost even more. These costs have to be passed on to the business customer.   In addition to rental costs, the service provider has to deploy an Internet Access Device on the customer premise, and often has to perform internal wiring and phone installation services, so there is a big labor and equipment cost element.  So, if you want to control service quality for your customers by renting somebody else's network, you are buying into a lot of extra work and extra cost.

However, when it comes to customer retention, the extra cost of renting a network and managing quality can be worth it.  Network renter Cbeyond's monthly customer churn in the 2nd quarter was 1.3%, while over-the-top provider 8x8 had churn of 3.5%.

There are several factors involved in the disparity between these churn numbers.  For one thing, buyers of managed services are often larger and more stable, so they don't go out of business as much as smaller businesses that buy over-the-top services.  Also, 8x8's 3.5% churn number includes churn in their shrinking legacy residential customer base, which is higher than the churn rate in their growing business customer base.  However, I think the bigger factor is that installing infrastructure at your customer's site, and then managing that service, can increase the stickiness of the your service, because your customer knows you, has met you, and your customer knows that replacing you involves a lot of work on everyone's part.   Over-the-top service providers can be ditched pretty quickly, though...the customer doesn't even need to change their Internet access to make a switch.

Does that mean that 8x8's over-the-top business services are doomed to face the same issues as residential VoIP services like Vonage?  More to come tomorrow. 

The Death of Over-The-Top Primary Line Residential VoIP

I fear we are witnessing the death of over-the-top primary line residential VoIP.  The largest player in the space, Vonage, is spending $65M per quarter just to keep their subscriber base from shrinking.  Another large player, 8x8, decided at the end of 2006 to focus on a business product, and while their residential subscriber counts temporarily benefitted from the 2007 SunRocket bankruptcy, 8x8's residential subscriber counts are now declining.  Could it be that in order to survive in primary line residential VoIP, you need to own your own network and offer the voice-video-Internet triple play? 

From the beginning of the residential VoIP primary-line-replacement market, upstart competitors like Vonage and 8x8 have emphasized cost savings as the primary attaction for subscribers, with a wink-wink understanding that service availability and audio quality may not match plain old telephone services.

Today, this basic value proposition is under assault:

• Cost-sensitive buyers are being siphoned off by ultra-low cost VoIP services like Skype, which don't even pretend to be a primary line replacement service. 
• Middle-market and quality-sensitive buyers are being enticed by the higher audio quality of "digital cable" voice services (which are really VoIP services), along with a total cost of service savings on a bundle of voice, video, and high-speed Internet services.

This leaves Vonage in a pickle, as they try to retain subscribers while simultaneously trying to create a new and better product suite, one capable of winning new subscribers away from the cable companies. 

The deeper question, though, is for 8x8.  The company has been successfully re-focusing on business VoIP services since 2006, making their loss of residential market share less painful.  But cable service providers are now rolling out business VoIP services of their own.  Will the cable companies come to own the small business VoIP space, much as they have come to own the residential VoIP space?  Can 8x8's Virtual Office offering compete against the cable companies?   

Vonage Improvement Not Enough To Escape Danger

Vonage reported second quarter earnings this morning, and posted better churn numbers than last quarter, down to 3.0% monthly churn now.  Churned subscribers nearly completely offset 230,832 gross subscriber additions, yielding a net gain of 2,080 subscribers for the quarter.  Last quarter, churn was at 3.3%.

We are seeing the effect of the churn ceiling in action, here.  At a 3% churn rate, and with average gross subscriber additions of 230,000 per quarter, Vonage will hover right around their current 2.6 million subscriber count, indefinitely.  And, if recent trends hold, it will cost Vonage agout $65M per quarter in marketing expenses just to keep their subscriber count at the 2.6 million level.

That's why it's so important for Vonage to either reduce churn more, or win more subscribers per marketing dollar, or both.  Vonage has pinned their hopes on new CEO Marc Lefar to get the job done, since he has a good record of churn-fighting at Cingular.  I hope he succeeds, but honestly, I'm skeptical that anybody can turn Vonage around and restore it to its high-growth glory days. At this point, mere survival is a win for Vonage.

8x8 Continues Successful Makeover

VoIP provider 8x8 announced earning yesterday, and it is clear that they are continuing a successful transition from consumer VoIP services to business VoIP services.  Revenue held steady at $16.3M last quarter, the same as the previous quarter, but the revenue mix continued to tilt more toward business services, so margins and earnings improved.

The company shifted their focus to business services in December, 2006, so we have had about a year and a half to see how the strategy shift is working.  The chart tells the story...consistent growth in the higher-margin business services, offset by a gradual shrinking of the consumer VoIP base. 

The chart below shows that business services now account for 56% of of overall 8x8 revenues:

Churn is also improving, though it is still too high at 3.5%. 

Kudos to 8x8 for recognizing early that the consumer VoIP market would become a competitive meat-grinder, dominated by the multi-service cable bundles.  8x8 was smart to shift to business services.

Vonage Names Churn-Fighter Lefar as CEO

Vonage named Marc Lefar as their new CEO yesterday, replacing founder and interim CEO Jeffrey Citron.  Citron will step back into a non-operating role as chairman of the board of directors. 

Lefar has a long history as a marketing executive, holding positions with Cingular Wireless, Cable & Wireless Global, Verizon Wireless, GTE Wireless, and Procter & Gamble.  He graduated from the University of Virginia in 1986.

Vonage hopes that Lefar's success in reducing churn at Cingular can be translated into similar results in customer retention at Vonage.  The company has struggled with churn rates above 3% per month. 

Skype Struggling?

Hudson Barton has an interesting post over on the Borderless Communicator blog, saying that Skype is struggling.  He points out that the percentage of registered Skype users that really use the product has been declining for years, and is now below 10%.  It might be a stretch to say that Skype is struggling, when it looks like revenue will grow more than 40% this year, but I understand what Mr. Barton is driving at.  While the number of registered Skype users continues to grow rapidly, fewer and fewer of the newbies are using the product on any kind of regular basis.  What gives?

Barton offers three theories:

• The isolated Skype development team can't generate enough features to keep people interested.
• Skype's marketing style might be flawed because they don't interact with their development community or with their user base very much. 
• Skype hasn't really provided much incentive for the hard core Skype users to evangelize the product.

All of this causes me to wonder:

• Is Skype out-growing its target market? Could it be that the kinds of folks who are loyal and frequent Skype users are a small subset of the population, and the remainder are just not ready to use a PC client as their phone device?  Yes, you can get Skype in handset form, either wireline or mobile, but how many people would even think of that?
• Could it be that most people prefer plain old mobile or wireline phone service versus a PC client?   

In any case, it looks like Skype has another chasm to cross in order to sustain a rapid growth rate. What do you think would help?

Vonage Refinances Debt, Trades One Question Mark For Another

Yesterday, Vonage announced that they had signed an agreement to refinance up to $215M in debt that is maturing on December 16th, 2008.   Silver Point Finance will be leading the deal.  This effectively removes a major cloud that has been hanging over the stock all year long.  However, the terms of the refinancing are scary, and as a result the stock is still stuck at about $1.60 or so. 

According to the Form 8-K that the company filed with the SEC, there are two parts to the refinancing - a senior secured debt facility (for $95M-$125M) and a traunch of senior convertible notes (for $90M).  Combined, these could provide $215M in proceeds to pay down the $253M in debt that matures in December.  The company presumably would use cash on hand to pay off the remaining $38M they owe on the maturing debt.  The company also has to pay between $21M and $24M in fees and expenses just to close the new debt round.  So, the company is burning $59M to $62M in cash to restructure its debt.  As of the company's first quarter report this year, they had only $148M in unrestricted cash to spend on this.  We'll find out how much cash the company has now on August 7th, when the company reports 2nd quarter earnings.

The variable interest rate on the new five-year senior secured debt facility starts at about 15% and could go higher based on the higher of LIBOR and the prime rate.  The company is also required to prepay some of this loan if assets are sold or if the company has more than $75M in unrestricted cash on hand.  Any mandatory or voluntary pre-payments will be encumbered with additional fees that jack up the effective interest rates to even higher levels.  These terms leave the company vulnerable to an increasing interest rate environment, and even if the company is successful in generating free cash flow, the company will be paying a big chunk of it to the senior debt holders for the forseeable future. 

The $90M convertible notes are on a 7-year amortization schedule and will pay an interest rate of 14%, but some of the interest in the first year is deferred.   The notes convert at the lesser of $2 per share or 110% of the 30-day average stock price prior to the "pricing date".  The minimum stock price for conversion is $1.65.  The company can require the conversion of these notes after three years.

The bottom line for the holders of Vonage stock is that for the next several years, even if the company performs well, the debt holders will reap the rewards instead of the stock holders, and there is the possibility that stock holders will be diluted to the tune of $90M at $1.65 to $2.00 stock prices.  If I'm reading it right, this dilution at $1.65 means about 54.5M new shares would be issued, versus the 156.1M shares currently outstanding.  The 8-K says something about full-ratchet anti-dilution provisions...I'm not sure what that means, so perhaps a reader of this blog can enlighten me.

In any case, it looks like Vonage is in debtor's prison for a few years and has to earn its way out.  Meanwhile, shareholders have a long wait to see much upside.  The company has traded the uncertainty of the debt refinancing for the certainty that the company must generate positive cash flow or perish at the hands of their lenders. 

   

Skype's 2nd Quarter Growth In-Line

EBay announced 2nd quarter earnings yesterday, and with the earnings they announced that Skype had achieved $136M in 2nd quarter revenue, right in line with the "realistic case" revenue forecast I put out back on March 8th.  Skype's subscriber growth is slightly behind my forecast, but their revenue-per-user is slightly ahead of forecast.  Here is how Skype has done the past three quarters:

SKYPE STATS	Q4 2007	Q1 2008	Q2 2008
Borderless Communicator "Real Users"*	27,206,646	31,741,318	32,225,000
Quarterly Revenue ($millions)	$         115	$         126	$         136
Quarterly Revenue Per User	$        4.23	$        3.97	$        4.22

The Borderless Communicator "Real Users" stat comes from Hudson Barton's Borderless Communicator blog, here.  Mr. Barton tracks how many of Skype's reported registered users really used the service on a regular basis, so his "Real Users" stat is a much more reliable indicator of how many people really use the service.  If I have one gripe with Skype, it's that they subbornly cling to reporting "registered users", when everybody knows that a good number of those registrations are dormant.  I myself have three Skype accounts, and only one of them ever gets used.

Overall, Skype's revenue and subscriber growth are slowing down in the 2nd quarter.  The same thing happened last summer, though, so I factored that into the forecast since this could just be a seasonal pattern.

Here's the March 8th forecast for 2008 for Skype:

	Q1 2008	Q2 2008	Q3 2008	Q4 2008
Quarterly Revenue per "Real User"	$        3.97	$        4.18	$        4.18	$        4.18
Total "Real Users"	31,741,318	32,608,275	33,110,482	36,410,482
Quarterly Skype Revenue ($M)	$         126	$         136	$         138	$         152

If the rest of they year plays out according to this forecast, with big jump in 4th quarter users, then Skype will be at about $553M in revenue for 2008.

Who's Winning the Cable-Telco Phone Wars?

For the past couple of days, we've seen how the big telcos are kicking cable's tail in the video market, with the telcos achieving organic growth at about five times the rate of cable.  So, cable is clearly losing the TV subscriber war. However, in the battle for phone subscribers, cable is taking the telcos to the cleaners.

In the year ending March 31, 2008, the two largest telcos, AT&T and Verizon, lost over 8 million access lines, while the four largest cable companies, Comcast, Time Warner Cable, Charter, and Cablevision, added over 4.5 million phone subscribers.  The table below tells the tale of woe for the telcos:

Phone Subscriber Additions/Contractions	Year Ending March 31, 2008
Telco Access Line Losses	(8,647,000)
Cable Phone Subscriber Growth	4,508,400

Of course, the telcos are replacing some of their losses with growth in wireless subscriptions, which is a point of strategic vulnerability for the cable companies.  However, it lookes like over half of the line losses experienced by the telcos are going to the cable companies.

Here's the detail of line losses and phone subscriber gains, by company:

Phone Subscribers	1Q07	2Q07	3Q07	4Q07	1Q08
AT&T Access Lines	65,429,000	64,078,000	62,871,000	61,582,000	60,415,000
Verizon Access Lines	44,154,000	43,293,000	42,410,000	41,441,000	40,521,000
Comcast Phone Subscribers	2,471,000	3,142,000	3,804,000	4,377,000	5,016,000
Time Warner Cable Phone Subscribers	2,094,000	2,335,000	2,610,000	2,900,000	3,180,000
Cablevision Subscribers	1,320,000	1,399,000	1,490,000	1,592,000	1,685,000
Charter Phone Subscribers	572,600	700,300	802,600	959,300	1,085,000

So, the cable companies are winning far more phone subscribers (4.5 million) than the telcos are winning TV subscribers (nearly 2 million), at the moment.  The cable companies would be happy to continue losing TV market share to the telcos at this rate, so long as they are gaining phone market share at this much greater rate. 

 

David Bryan Does A P2P SIP Interview on SquawkBox

Regular readers of this blog will know of my interest in peer-to-peer Session Initiation Protocol (P2P SIP), and my optimism that the developing technology will form the next leg in the evolution of Voice over IP.  I did a print interview with David Bryan, CEO of SIPeerior Technologies and co-chair of the IETF P2P SIP working group, earlier this year to explore the technology.

Now, Mr. Bryan has done a great audio interview on Alec Saunders' regular SquawkBox feature, spending the better part of an hour talking with Dan York about peer-to-peer SIP and its applications.  It's a great interview, covering a lot of ground, including P2P security and how a P2P endpoint joins a P2P cloud.  If you are interested in learning more about what may become the next thing for VoIP, it's an interview worth your time.

Vonage Skid Continues Amid Debt Uncertainty

Vonage missed a commitment to refinance their $253M in convertible debt in the 2nd quarter, and as a result, the company's stock price is losing ground, hitting a year-to-date low closing price of $1.55 per share on Monday.

The convertible debt must be paid back in December, so the company has only five months of runway left to resolve the issue.  Most of the $253M must be refinanced because the company does not have enough cash on hand to pay it off, and because the company does not generate much free cash flow from operations. 

Vonage signed a non-binding letter of intent with a financier in April to refinance the debt, saying that two-thirds of the financing should come from a senior secured credit facility, and one-third from new convertible debt.  In May, CEO Jeffrey Citron said that he expected the transaction to close in the 2nd quarter.  The company missed a June 13th deadline to finish the deal, and then missed the extended deadline set for June 23rd.

Now the company hasn't set any public deadlines. 

Without a refinancing, the options are slim:

• Sell the company to a public or private buyer, or
• Declare bankruptcy

I don't think there is public company that wants to buy the company and assume the debt, so if the debt can't be refinanced in this tough market, maybe the company would declare bankrupty to erase the debt, and then be sold. 

Of course, if Vonage can complete the debt refinancing, then they get to live another day.  However, every day that goes by without debt refinancing increases the chance of bankruptcy.  And the stock price shows it:

   

QoS Strategies of Business VoIP Providers

For the past few days I've had a series of posts showing the results from an informal survey of Business VoIP service providers.  Wednesday, we looked at size rankings by number of seats served, and yesterday we looked at which providers sell retail, wholesale, or both, and the geographies served by each.  Today, we'll look at quality of service strategies of the various providers.

As you can see from the table, all seven of the pure retail providers actively manage the quality of service provided for their retail customers.  In order to do that right, you generally also have to sell Internet access to the customer and you have to provision a circuit to the customer site, connected to a managed router on the customer end, and connected to your own network-side router on the other end. Then, you configure the routers on either end of the dedicated circuit to prioritize VoIP packets over other plain data packets, so that voice quality isn't disrupted by a burst of data traffic.

The hybrid wholesale/retail providers are more of a mixed bag.  8x8 and Junction Networks do not actively manage the quality of their VoIP services, preferring to rely on the public Internet for their customer connectivity.  This gives 8x8 and Junction Networks easy access to a national footprint, but it also means that customers may sometimes suffer from poor quality if their Internet connections are either overutilized or low capacity. 

The other hybrid wholesale/retail providers, Bandwidth.com, Simple Signal, CommX, and PBX-change, offer some degree of quality of service management, at least in some circumstances.

The wholesale players, such as New Global Telecom, often rely on their service provider customers to provide the Internet access link to the retail customer, so quality of service management is sometimes incomplete.  Unless the wholesaler controls the routers on both the customer premise and on the network side of the customer connection, or asks the retail service provider to do the same, quality of service cannot be guaranteed.  With that in mind, I'd be interested to learn how Vox Communications, a wholesaler, carefully manages quality for "TDM-like voice quality".  It might be that Vox is relying only on an adaptive codec, like Skype. 

      

Wholesale and Retail Business VoIP Providers

This week, I'm publishing the results of an informal survey of fourteen business VoIP service providers.  Yesterday, I published the list of survey respondents ranked by number of seats served.  Today, I'll look at the breakout of which companies are wholesale, retail, or both, along with whether the companies are focused in specific geographic areas, or are focused nationally.

The chart below shows the results.  You can click on it to see a bigger and better resolution version of the chart.

The size of the bubble indicates the relative size for the company, based on number of business VoIP seats served.  If a bubble straddles the wholesale/retail line, then the company serves both wholesale and retail customers.

The main take-away from the chart is that all of the pure retail companies, led by CBeyond, are focused on selected cities or are focused regionally.  That makes sense because it often takes a local sales presence for retail sales.

The pure retail players are led by CBeyond, followed by M5 Networks, Zayo Managed Services, and Unity Business Networks, Versature and Vivid IP - LogiSIP.  Fusion did not report their size, so the "proper size" of their bubble isn't known.  PBX-Change, a Tampa-based wholesale/retail player operating in the southeast, also did not report their size.

There are two pure wholesale companies that responded to the survey, led by New Global Telecom, at 82,000 seats served. Vox Communications did not report their size so the proper size of their bubble is not known.  Of course, their are other pure wholesale companies in the market, but they just haven't responded to the survey, yet.

The hybrid wholesale/retail players are lead by Bandwidth.com, followed by 8x8, Junction Networks, Simple Signal, and CommX. 

More data to come tomorrow! 

Business VoIP Survey, July, 2008

It's time to publish the first Telecosm Business VoIP Survey, in which Business VoIP companies self-report about the rapidly growing Business VoIP segment of the market.  This is a little experiment I started about a month ago, when I published a blog post asking Business VoIP companies to give us a little more insight into their business. 

This market segment is served by many regional players and many of them are privately held, so it's sometimes hard to get a handle on how the segment is growing.  I hope this little survey will help to shine a little light on the industry.  This should be the first of many periodic editions of this survey, and I hope that other business VoIP providers will join in.  I am very thankful for the fourteen companies that did respond.

There were fourteen responses to the survey, but since CBeyond is public and an ackowledged leader in the space, I added them to the list as if they had responded to the survey.  Fifteen companies does not make a very comprehensive industry survey, but it's a good start and begins to paint the picture of this emerging market segment, and I expect that the publication of the first survey will attract the attention of other companies in the space and the survey will grow as I publish it again each quarter. 

Here are the size rankings of the thirteen companies in the report:

After CBeyond, it looks like Bandwidth.com and 8x8 are the next-largest challengers, at a about a quarter the size of CBeyond.  Junction Networks and M5 Networks are next, followed by a batch of mid-sized providers: Simple Signal, Zayo Managed Services, CommX, and Unity Business Networks. 

I'll publish excerpts from the survey over the next couple of days and will publish the entire survey on Friday. 

Here is the contact information for each of the survey respondents:

Note: you can see larger versions of either of the images in this post by clicking on the image.

Final Call for Business VoIP Provider Survey Responses

Back in early June, I put out a call for Business VoIP service providers to respond to a survey, and promised to publish the results in early July.  Well, the time has come, and tomorrow is the day to publish the results.  It's not too late to get your response in...I just need to receive it by the end of the day today.  Just follow the instructions on the June 6th blog post.

So far, I've got responses from:

• 8x8
• Fusion
• M5 Networks
• Simple Signal
• Unity Business Networks
• Versature
• Vivid IP - LogiSIP
• Zayo Managed Services

I'll also throw in data about CBeyond, gleaned from their public reports.    Nine entries in the table is a good start, but I'd love to see a few more service providers participating in the survey, so spread the word and ask your service provider to lob in their survey response!

Skype 4.0 For Windows Places Bigger Bet on Video

Skype launched a beta version of their new Skype 4.0 for Windows software this morning with a worldwide conference call.  Skype calls the new 4.0 beta a "major redesign" of the user interface to try to make the client software easier to use, and to upgrade the video quality in the client. 

What's new? The most impressive work centers on improved video quality.  The video capability is truly outstanding, taking advantage of a larger window size, and an improved user interface lets you start a video call with a single click.  Skype uses an adaptive video codec that adjusts the amount of bandwidth used to the capacity of your Internet connection, so a video call can use as little as 48 Kbps, but the video really starts to look good at about 300 Kbps.  Skype is clearly trying to ride the momentum of a growing base of video callers, saying that the percentage of calls carrying video has grown from 8% three years ago to 28% today. 

One of the questions on the conference call asked "Why do you think video will take off with VoIP, when it clearly hasn't with 3G video calling?"  Michael Bartlett, Director of Windows Product Management at Skype, answered by saying that the mobile phone form factor doesn't lend itself to video calling as well as a desktop or laptop computer, and that Skype's own data show that video is taking off, again citing that 28% of calls carry video.  They clearly expect that percentage to grow.

The new software has also made a number of other improvements to the user interface, also taking advantage of greater screen real estate.  It is now easier to find friends or contacts, easier to start and manage conversations, and easier to share files with each other in a conversation.  Skype is eating deeper into the territory of Webex and the other web collaboration players.  

Skype hasn't set a hard date for launching the final public version, saying they really want to listen to users during the beta testing to "make sure this is the best piece of software we've launched in our history."

Skype says they "anticipate some criticism" as users adjust to the new software.  First impressions are good, though, and in my view it represents another step forward for Internet video in general, and for Skype in particular.

Cheap VoIP Service: iCall on the new iPhone

With yesterday's launch of the new 3G-capable Apple iPhones, VoIP over WiFi is getting a boost.  I still believe using VoIP over WiFi will be often be plagued with quality problems, but if you are really interested in conserving your cell phone minutes, you might want to check out a new iPhone app called iCall. 

iCall is a free VoIP over WiFi application for the iPhone, and you can download a beta version of it now.  Basically, it's a tool to save minutes if you are on a limited minutes cell phone plan.  Whenever you are within range of a WiFi network, you can elect to make outbound calls over the WiFi network instead of AT&T's mobile network, and you can elect to transfer calls that you have received on AT&T's mobile network over to the WiFi network.  When the call is on the WiFi network and not on AT&T's network, then you are not eating minutes from your cell phone plan's minute bucket.  

This is a lot like the T-Mobile "Hotspot @Home Talk Forever Mobile" service that won a BUNGL few months ago, except that it's free.  (That was probably the most controversial of the BUNGL awards...took a lot of heat for that one!  Needless to say, iCall won't be winning a BUNGL.)

How can iCall afford to give away domestic long distance and local phone service?  Good question.  The web site says this about that:

"Why would we give this to you for free? The answer is simple - iCall wants you to call the businesses who pay to be listed in our directory, and we'd also love to sell you a headset or dirt cheap international calling rates."

So, it's an ad play, supplemented by upselling on international long distance and accessory sales.  I hope they generate enough revenue from that to cover their costs, or there are going to be a lot of disappointed iCall users!

More on Magic Jack

Coincidentally, Doug Mohney over at FierceVoIP also posted on Magic Jack today.  He added a bit more detail to the business plan, that I didn't cover in my post: the company plans to sell advertising and reserves the right to tailor ads based on the phone number that you dial.

According to Mohney's post, Magic Jack is adding users at a rate of over 7,000 a day, faster than Vonage! 

Cheap VoIP Option: Magic Jack

My friend Michael Remacle, who now runs business development for Vail Systems, suggested I do a post on Magic Jack, one of the truly inexpensive VoIP options out there for consumers.  No, I'm not talking about Level 3 CTO Jack Waters...though I understand how you might think that!  No, Magic Jack is a separate company that sells a really cheap consumer VoIP device and service.

Here's how Magic Jack works:

1. You buy a USB "magic jack" dongle for $40, which includes all of the fees for your first year of service.
2. You plug the USB dongle into a USB port on your PC.
3. You plug your standard, traditional analog phone set into the phone jack on the USB dongle.
4. You start to make and receive unlimited VoIP calls to and from any regular telephone.
5. You also get free voice mail.

So, it's basically $40 for your first year of service (including the USB dongle), and $20 a year for every year thereafter.  This computes to about $3.34/mo in the first year and about $1.67/mo for every year thereafter.  Cheaper than chotskis at a VON conference!

I have never used the service so I can't vouch for it.  It certainly is low-cost, though!

How can anybody afford to sell VoIP service so cheap?  After all, if you assume an average of 500 minutes of long distance per user, per month, then in the second year they are basically charging $1.67/500 = $0.00333 per minute of use.  That's well below current wholesale long distance prices, let alone retail pricing, so you do have to wonder how Magic Jack is selling phone service for so little.

Here are some theories on how Magic Jack can sell so cheap:

1. They are enjoying interest on the float, since they are collecting annual fees up front.
2. They are benefiting from some breakage from folks who buy the service but end up not really using it (so the average long distance minutes used per month is actually lower than 500 minutes).
3. They are gaming the reciprocal compensation system to get paid for some of their inbound calls (more on reciprocal compensation in a future post).
4. The cost of goods on their magic jack dongle is really tiny, like less than $5.00, so if you attribute $20 of the intial $40 payment as the retail price for the jack itself, then they are making really healthy margins on the cheap hardware.  (It would have been great to have a sub-$5.00 dongle 9 years ago...this would have come in real handy when we were searching for cheap bricks back then!)

Even with these benefits, though, it's hard to see them making much money at it.  Just another example of how it's hard to compete as a consumer VoIP service provider these days.  It's a good deal for consumers, though!   

VoIP Business Survey Responses Starting to Roll In

On Friday, I put out the call for responses from all business VoIP service providers to participate in a survey of this emerging space. The responses are starting to roll in, thanks to some help in spreading the word from Peter Radizeski at RAD-INFO, and from Rob Powell over at Telecom Ramblings. Many thanks to those who have already responded to the survey, including VoxBone, Vivid IP - LogiSIP, and Versature.

The survey responses will be published in early July. Please help to spread the word...the more responses we get, the better picture everyone gets!

Calling All Business VoIP Providers

A few days ago I posted a small roundup of business VoIP service providers (see "Four Different VoIP Business Models"), and I have had a bunch of comments and emails about that post, wondering why I didn't list any of a number of other examples of business VoIP service providers in that post.  The truth is, business VoIP providers are thicker than summertime flies in Maine right now, and there are many more than four different variations of VoIP business models.  Most of these companies are privately held, so they are harder to track than public companies. 

There is a real lack of comprehensive information about these companies, so I am proposing that we keep a public listing of these companies in a quarterly posting on this blog.  This could be a win-win for everybody:

1. The VoIP company gets a little more publicity, and I'll post email/phone/website contact information for each company on the list.
2. The blog's readers, the VoIP companies, and I get a more accurate view of how this market is developing.
3. I get to put interesting stuff up on my blog!

Here is how I propose it should work:

• If you are a VoIP company that focuses on serving businesses, email me (at ike@practiq.com) all the info you are willing to make public, hopefully including:
  1. company name,
  2. contact phone,
  3. contact email,
  4. web site URL,
  5. number of seats served at the end of your most recent quarter,
  6. Sales strategy: wholesale/retail/both,
  7. Sales geography: selected cities/national/international (list specific geographic focus if you wish)
  8. Distribution strategy: feet on street/telemarketing/retail stores/indirect channels (agents, etc.)
  9. QoS strategy: managed/unmanaged
  10. Do you sell Internet Access?
  11. Do you sell a service that allows your customer to run their own traditional PBX (often called a "trunk replacement" service)?
  12. Do you sell a service that allows your customer to run their own IP PBX (often called a "SIP trunking" service)?
  13. Do you sell local and long distance minutes?
  14. Do you sell hosted IP PBX services?  If so, do you own and operate your own IP PBX, or do you purchase acess to a wholesale IP PBX service?
  15. What is your average number of employees per customer site?
• Any info you don't want to share I'll just list as "DNR", or Did Not Report.
• I'll compile all of the responses I get in tabular form and post it once per quarter.  First posting will be in early July so the deadline for responses for the first posting will be June 26th. 
• I'll also do interviews with executives from this company list from time to time, and post them on the blog.

So, submit your info, and we'll get going on this!  If you are a Telecosm reader and know of companies who may want to participate in this survey, you can help by sending a link to this post to your contacts, so we can get as comprehensive a view of this market as possible.

Thanks!

   

Why SIP Hasn't Won Yet

Brough Turner has a wonderful, succinct summary of why the Session Initiation Protocol (SIP) hasn't realized its potential, at least not yet.  Brough and I agree that peer-to-peer SIP is the best chance for the protocol to completely change communications as we know it.  Check it out!

Four Different VoIP Business Models

My Grandad had a bunch of wise sayings, and just being around him, you could be guaranteed of hearing at least one per day.  One of his favorites was "It takes all kinds of people to make a world," and he would trot that one out whenever somebody did something strange, or whenever somebody issued a broad-brush complaint about a group of people. 

You can extend that to VoIP, too, where it takes all kinds of companies to make a VoIP world.  There's no guarantee that all of the different VoIP service provider experiments now underway will win out in the end, but there are a lot of experiments now underway.  We've looked a 8x8 and Vonage in depth over the past couple of weeks, and those two represent companies focused on the small end of the market:  residential and SOHO customers.  What are some the companies focused a little higher on the VoIP pyramid? 

The table below shows a half-dozen VoIP-based service providers, roughly rank-ordered by the number of extensions at the average customer site.  Vonage is at the bottom, because as a residential service provider they usually sell about 1 line at a time.   

Next up the list is 8x8's Virtual Office service, which averages less than five employee extensions per customer site. 

CBeyond averages about 7 extensions per customer site, and they differ from 8x8 in some significant ways.  For one thing, CBeyond is a CLEC and they re-sell local exchange carrier capacity to the customer, usually in T1 form.  CBeyond also does Quality of Service (QoS) management of that pipe so that voice packets have higher priority than other data packets.  They then sell both Internet access and voice services on that converged T1 pipe.  Lastly, CBeyond also does something unique among the service providers on this list:  they re-sell a wireless phone service and can bundle that into the service offering.  8x8 does none of these things, but 8x8 does offer a hosted IP PBX, while CBeyond does not focus on a hosted IP PBX offering.

Next up on the scale are Unity Business Networks, M5 Networks, and LightEdge Solutions.  All three of these companies are privately held, so there is little public information on these companies, and as a result, some of the data in the table should be considered an "educated guess".  For example, I'm guessing all three of these companies average greater than 10 extensions per customer site.  Part of the reason for that is that unlike CBeyond, these three private companies sell a full hosted IP PBX solution to their customers, while CBeyond does not. 

All of the top four service providers on the list use a "feet on the street" sales force.  Most of them use agents to sell their product as well.  All of the top four service providers have focused their sales efforts on selected cities, rather than attempting a national or international marketing campaign.   This has the disadvantage of reducing their market reach, but has the big advantage of helping to build local brand awareness and also helps to build deeper, more lasting relationships with customers and community leaders.

Different approaches should yield different results.  For example, I believe that if you go further up the customer size chart, and combine that with a geographically-focused, feet-on-the-street sales approach, you should see lower churn than the national marketing machines that sell to smaller customers.  So far, the data on the public companies is bearing that out, with CBeyond showing churn in the mid-2% range, and 8x8 showing churn in the mid 3% range for their business services.

Full disclosure: I am chairman of Unity Business Network's board of directors.    

Should Vonage Copy 8x8?

For the past several days we've looked at how 8x8 is successfully converting itself from a residential VoIP focus to a business VoIP focus, and is seeing improving profitability as a result.  So, what does this mean for Vonage?  Is there hope that Vonage can re-invent itself as well?

I don't think the simply copying 8x8 will rescue Vonage.  For one thing, 8x8 is less than one tenth the size of Vonage, small enough that it only took 8x8 about 18 months of new focus to have business revenues reach half of total revenues.  If Vonage were to try the same thing, it would take much, much longer for Vonage's SOHO revenues to match their residential revenues, and SOHO revenues might never reach those levels. 

Then there is the question of whether Vonage can achieve the same economics as 8x8 in a business play.  Part of why 8x8 is profitable in their SOHO market is that they wrote a lot of the IP PBX software themselves, so they don't have to pay as much in third-party licensing fees as competitors who use off-the-shelf software like Broadsoft.  Does Vonage have the software to offer the full suite of IP PBX features that businesses want?  If not, the economics of a big Vonage SOHO push will look a little worse than 8x8's. 

Intellectual property has to be a consideration, too, especially since Vonage has attracted patent lawsuits like teenage boys to a cage-fight.  Since Vonage probably doesn't have any IP PBX patents, they would need to license the technology from somebody who can indemnify them against lawsuits. 

So, while Vonage could dive into a SOHO offering in a bigger way, it is unlikely to generate complete replacement revenue for their residential business, and is unlikely to be the silver bullet that it has been for 8x8.  A ramped-up SOHO effort might help some, though, in combination with other initiatives.  It just won't be the complete answer by itself.

8x8 and the Double-Edged Distribution Sword

Yesterday, we looked at how 8x8's Virtual Office hosted IP PBX product is having its greatest success in the SOHO (Small Office/Home Office) segment of the business market, and how the product would have its greatest appeal to that segment of the market.  Today, I'm looking at 8x8's distribution strategy and how the company's distribution methods also target the low end of the business market.

8x8's biggest sales channel is an inside sales team, tasked with taking calls and emails from interested customers and close deals.  The company also sells through big-box retailers like Office Depot, and has a trial relationship going with Staples as well.  Lastly, the company sells some of their services through outsourced telemarketing and through agent channels.  Based upon data from the company, in the first quarter of this year, here were their sales sources:

8x8 relies on a combination of advertising, lead generation, and distribution partnerships to generate inbound calls into the company's sales team.  The company has been very active in trying new marketing techniques, and the trial and error has likely generated quite a bit of tribal knowledge about what works and what does not.

The company's distribution strategy has a couple of great advantages:

1. An inside sales team is less expensive than a national sales force, so the cost of sales is lower, enabling the sale of a service with a lower price tag.
2. An inside sales team can achieve greater geographic reach more quickly than a "feet on the street" approach.

However, the inside sales approach is likely to be most effective in reaching SOHO businesses.  Call me old-fashioned, but I believe most businesses with a larger number of employees will want to see the face of a salesperson, and will want to hear first-hand that their needs will be met.  Not so many will be willing to buy on the basis of an online ad and a phone call. 

That's not to say that larger companies don't buy 8x8's service.  The company boasts some national names among its customer list, including Subway, H&R Block, and RE/Max.  For the most part, though, these companies have branch offices with a very small number of employees in each office, and for a very small office, a Virtual Office solution over third-party DSL is usually going to be less expensive than a managed T1 solution with a hosted Broadsoft service.

To reach more deeply into the ranks of the larger offices, 8x8 would either need to rely more heavily on an agent channel, or put their own feet on the street, in the style of Unity Business Networks and CBeyond.  However, the feet on the street approach is more expensive and doesn't cover as much geography as quickly, so for now, I expect 8x8 to continue to grow with its inside sales approach.  It will be interesting to watch their growth trajectory versus CBeyond's.