Mercator Capital is a privately-held investment bank focused on mergers & acquisitions, private placements, and strategic advisory services. Mercator's IP Communications Newsletter is a monthly analysis and commentary on the major business stories impacting the convergence of voice, video, data, and wireless communications.

1. Acme Packet IPO – No Friday The 13th Jitters Here

On Friday, October 13, Acme Packet’s conducted its IPO, and by all accounts it was hugely successful. Their public filing was announced on June 2, not long after Vonage’s ill-fated IPO, which naturally raised some concerns about the timing and viability of Acme’s plans. Many had speculated that the Vonage IPO had tainted any new offerings that were connected to the VoIP industry. Although Acme is much smaller, less well-known and in a very different space than Vonage, investors rightly chose to evaluate them on their own merits.

Though the original range was $6.50-$7.50 per share, the initial price was raised 46% to $9.50 before trading began on opening day. This was an indication of strong demand for the stock, which was further validated by their opening day performance, as well as aftermarket trading to date. Acme’s first day of trading closed at $15.91, an impressive 67% rise. Since then, the stock has hit highs north of $19, and has been holding its own in the $18 range, which translates into a market capitalization of over $1.1 billion. This is a very strong performance for a company that reported revenues of just $38 million during the first half of this year, and a very different story from the Vonage IPO debacle, despite Vonage having revenues many times the size of Acme's. Let’s take a look at some of the key differences between the companies.

First and foremost, Acme Packet is in a different business than Vonage – although they would be an ideal customer for Acme’s session border controller (SBC) solutions. Network infrastructure is not generally considered a sexy category, and on top of that, SBCs are complex and not well understood, especially among investors. SBCs were virtually unknown three years ago, and much the way Vonage has created market awareness of VoIP, Acme has probably done more than any other vendor to give credence to SBCs as a distinct product category.

Acme is truly a pioneer in this space, and by continually improving and evolving their technology, they have emerged as a market leader, and are the dominant SBC vendor among Tier 1 carriers around the world. They have over 300 customers across 70 countries, 91 of which Acme contends are Tier 1s. This includes 21 of the top 25 carriers, as well as four of the top seven cable MSOs.

Acme also remains one of the few independent SBC vendors following the much-expected consolidation in the sector. Mercator has covered these deals in previous issues, and with the recent exits of vendors such as Netrake, Jasomi and Kagoor, Acme has weathered the storm well. To their credit, Acme has done it the old fashioned way, by keeping costs under control and growing the business organically. Since inception, they have only taken $30 million in venture funding, and have been self-funding for the past three years. Again, this is a very different scenario from Vonage and the market in which they compete.

From a financial standpoint, Acme’s recent success is stunning. In 2005 they had about $36 million in revenues, which more than doubled to over $38 million during the first half of 2006. They maintain gross margins of 79%, and during the first half of the year generated $17 million in cash flow. Prior to the IPO, Acme already had $33 million in cash, and brought in over $80 million more after expenses through their offering. Compare this to Vonage who was basically spending more in marketing than the revenue they were bringing in, and you can see why the financial community embraced the offering. But even after backing out the current cash balance, Acme’s enterprise value is approaching $1 billion, which represents a multiple of 17X the company’s trailing 12 month revenues of $57.5 million. This is a lofty valuation by any standard.

There are a number of reasons why Acme has been successful – and profitable – and we will touch on a few here, as we see them as a blueprint for other vendors to follow in their footsteps. Veraz Networks comes to mind with their recently filed IPO, and others such as BroadSoft and NexTone are no doubt thinking along the same lines. The overall message is clear – to be successful, companies must keep innovating and moving forward, focus on big market trends, but also focus on revenue growth and above all profitability. This sounds very basic, but Acme has embodied these ideas as well as any vendor we have seen.

Most recently, Acme had a very strong showing last month at GMI 2006, a global IMS interoperability event. Their Net-Net product family demonstrated interoperability across all testing scenarios in four different test sites. In time, IMS will be the primary reference architecture for all carrier IP networks, and the ability to support multi-vendor solutions across all types of media will be a key table stake. The experience gained from deploying with so many Tier 1s globally goes a long way to making Acme strong in this regard.

VoIP peering is an important facet of what SBCs do, and carriers are starting to embrace peering as a means of getting more value out of their IP networks. As with IMS, this trend will continue, and eventually all traffic will use peering, which should bode well for Acme. To position themselves on this front, Acme has hedged its bets by aligning with all the key players. In September, they announced being certified as XConnect ready, and are now part of the rapidly growing XConnect Alliance. In June, they announced similar news with Stealth Communications, and now support the company’s Voice Peering Fabric. Acme has also partnered with Nominum and Siemens for their own peering solution.

Another value-added initiative is their Technical Certification Program, which was launched in June. With a large, growing customer base, and a new, complex technology, this program allows Acme to set industry standards where little expertise previously existed. This is a sound way to ensure that carriers are well-served and comfortable with Acme’s expertise in deploying their systems – not just for Acme personnel but their technology partners and systems integrators.

Acme has also moved forward on a few fronts to address emerging growth areas. First, they recently adjusted their philosophy to embrace both schools of thought on SBC architecture. Acme has long insisted that SBCs work best as an integrated solution, namely in terms of the signaling and media control functions being housed in the same system. This model is well-suited to Tier 1s, but is not as practical or economical for smaller carriers. Those carriers are more likely to adopt the disaggregated – or “decomposed” as Acme calls it – architecture, where the signaling and media control are separate and reside in different areas of the network. This may seem down-market to Acme, but the reality is that smaller carriers are now seeing the need for SBCs, and to address this market, it only makes sense for Acme to offer a decomposed solution.

Along these lines, another emerging growth market for SBCs is the rural carriers, namely IOCs. To serve rural carriers in the U.S., most vendors seek RUS designation, and Acme achieved this in August. In brief, IOCs can obtain low-interest government loans to finance network upgrades from RUS designated vendors. Without this financing, it becomes much more difficult for IOCs to justify the investment in IP network infrastructure.

Clearly Acme Packet is doing a lot of things right. We see them being well positioned for growth among all sizes of carriers, and they have the resources now to remain a pure play SBC provider for some time to come. It is very likely that the expected migration of SBC functionality to other network elements such as routers and media gateways will continue. However, we do not feel that these options can fully displace Acme, and nor do we see this coming from the legacy telecom vendors, who have yet to develop their own SBC solutions that approach what Acme has today.

On the other hand, we do expect to see downward pricing pressure as vendors chase Acme’s business, but this impacts everyone, and not just Acme. Could they become a target for a legacy vendor like Nortel or a data vendor like Cisco? Absolutely. Could Acme become a consolidator like AudioCodes, and extend their expertise beyond the SBC realm, perhaps into areas such as mobile peering, video peering or higher end security? This is also a very legitimate strategy in a VoIP equipment market that is seeking leadership – especially after the recent decision by Tekelec to divest their VoIP switching business after a failed attempt at building a portfolio through acquisitions.

With the proceeds from their IPO, Acme has a lot more flexibility to respond to the competition and opportunities before them. Based on what we know of the core management team, we believe they will continue to make the right calls.

Now that they are a public company, we can continue to watch their progress – they just announced third quarter results after the bell. The third quarter looks like a continuation of the great success they had earlier in 2006, as they reported revenues of $22.3M and net income of $6.8M for the quarter, while maintaining gross margins in the 79% range. Despite the great results, the company's stock was trading down after hours, possibly on lower than expected revenue guidance for Q4 and into 2007. We'll continue to watch their results and progress.

2. Cisco’s TelePresence Solution – Redefining Videoconferencing

On October 23, Cisco’s much-publicized TelePresence solution was released. Cisco reportedly spent a billion dollars developing the platform, and with that kind of investment, it is no surprise that the world knew about it by the end of the day. There is a lot at stake here, as Cisco is determined to make a name for itself in a space long dominated by incumbents like Polycom and Tandberg. On the other hand, enterprise is Cisco’s main domain, and they have certainly been successful going up against the legacy PBX vendors using their CallManager platform. But while this may be true, we have some reservations about Cisco’s ability to do the same with videoconferencing.

Since the early part of this year when Cisco brought forward their Unified Communications platform, they have had a strong focus on moving well beyond voice in the enterprise arena. This is very much a multimedia story, leveraging SIP and presence to make Cisco’s platform the centerpiece of all enterprise communications. In that context, video is a logical extension, but not solely as a desktop application. That’s where TelePresence comes in. Instead of going head-to-head against the likes of Polycom and Tandberg with a conventional system, Cisco is attempting to redefine the videoconferencing experience with an entirely different type of solution that puts them in a new category altogether.

TelePresence is a high-end solution that delivers an experience so life-like and life-size that the term videoconferencing seems almost out of place. From most accounts, the experience is very impressive and may serve as a viable substitute for face-to-face meetings. This is a critical part of their strategy since the driver for using TelePresence is to cut down on costly travel and to make remote workers more productive.

While Cisco should be able to deliver on some of these promises, the price of admission is high, and this is where some questions come to mind. First, at $299,000, TelePresence is orders of magnitude more expensive than most videoconferencing systems, even those that are high definition. As such, the initial market will be limited to Fortune 500 companies, so this is very much a niche offering. Secondly, TelePresence will appeal mainly to pro-Cisco environments, which already have significant Cisco infrastructure in place.

The system is proprietary and requires integration with CallManager. It also requires more bandwidth than most networks can provide, so many buyers will need to make network upgrades. They will also need to create a dedicated conference room to host TelePresence, and the entire environment must be customized and furnished by Cisco, right down to the color scheme of the room. In this regard, Cisco is going against the industry trend of open, flexible, standards-based systems, and has to be a cause for concern. There are many other lower cost, vendor-neutral solutions available, and if prospective buyers can get beyond the wow factor of TelePresence, they should give a long, hard look at the ROI and TCO metrics involved.

This brings us to a look at the competitive environment. TelePresence may be a high end, category-defining solution, but it is not the most expensive system on the market. That distinction goes to Hewlett Packard, and their Halo solution. Like TelePresence, this system was years in the making, and has very interesting roots. Halo was developed in partnership with the DreamWorks animation studio and delivers the same type of experience as TelePresence. Given the need for this type of system to be as life-like as possible, HP probably could not have picked a better partner.

Launched in late 2005, Halo was priced at $550,000 per room, along with an $18,000 monthly maintenance fee per room. The price has since dropped to $425,000, still above the price of TelePresence system, so Cisco may not need to worry about pricing themselves out of the market. As of September, HP had sold over 40 systems, so demand does exist for such an exotic system. A key element to HP’s success is the proprietary network they have purpose-built to support Halo, and in large part accounts for the high maintenance fees. By comparison, the monthly maintenance fees for TelePresence are only $3,500.

Even at Cisco’s lofty prices, they will have to sell thousands of systems just to generate $1 billion in revenue, let alone the profit required to pay for their huge investment. Clearly they are betting on product pull-through for other voice and data products to help provide a positive return. The general rule of thumb is that every dollar of voice products helps Cisco sell three dollars of data products; so the numbers are probably similar, if not greater, for video.

Interestingly, the high end of this market is being served by vendors outside the traditional videoconferencing space – HP and Cisco – and it is too early to tell if they know something that the rest of us are missing. Polycom and Tandberg are the resident incumbents, and have large installed bases, along with complete product families to address all levels of need. Polycom’s VSX product family is well established, but also serves the high end of the market with more recent offerings. In May, they launched RPX (RealPresence Experience), and with a price point of $249,000, have been addressing this market five months ahead of Cisco’s launch. Adding to this, in October their HDX 9000 series was launched, which provides a high definition solution at a much lower price point, in the $20,000 range.

Tandberg also brings a broad set of offerings, built around their MXP product line. The Profile series is comparable to Polycom’s VSX and HDX solutions, and their Centric and Edge systems address smaller videoconferencing requirements. LifeSize is less well known, but since coming to market in December 2005, they have shipped over 1,000 systems. They offer two solutions, but neither competes directly with TelePresence. LifeSize Room is their top HD system, selling for $12,000, and the scaled-down LifeSize Team sells for $8,000.

Given the rapid proliferation of PC-based video content in the past year, this discussion cannot overlook the PC as an important endpoint in the videoconferencing environment. TelePresence may deliver a life-size experience for select users, but PC-based video is becoming a practical alternative among a much larger user base. Two offerings in particular bear mention here. Tandberg, building on its long history in videoconferencing, launched a product called Movi in October. Movi is their PC-based solution, and brings cost-effective videoconferencing to small companies, as well as extending the experience beyond the boardroom.

SightSpeed takes this one step further with a free PC-based video calling platform. PC Magazine rated their 5.0 version “hands down” best-in-class last month. SightSpeed may not compete directly with TelePresence, or even enterprise-grade videoconferencing for that matter, but it is a proof point for how far IP-based video has come, and we feel this needs to be factored into the overall value proposition.

3. BroadSoft and Sylantro – Application Server Market Update

The application server market has been steadily evolving for some time, and in North America most of the vendor traction has been providing hosted PBX or IP Centrex-type platforms to carriers, primarily CLECs and Tier 2/3 operators. While this may be an oversimplification, there is little doubt about who the market leaders are – BroadSoft and Sylantro.

Interestingly, both companies held their annual partner and customer conferences the same week, from October 15-18, so we felt this would be a good time to assess the state of the sector. Clearly, momentum is strong for these companies, and having had first-hand accounts of both events, we have a better sense now as to why, and what the future prospects look like for both.

BroadSoft’s Connections 2006 event was held at the Phoenician Resort in Scottsdale, and Sylantro’s Global Summit 2006 took place at the Venetian Resort in Las Vegas. While it is not clear if the timing of these events was coincidence or by design, they provided a good window into the state of this sector. As vendor-sponsored customer events go, there were many similarities between them. Both were held at high end resorts where attendees were well taken care of, and the programming was carefully crafted to make customers feel good about doing business with them, and for prospects to be impressed by what they have to offer as well as the good company they keep.

The content for each event was engaging, providing attendees with lots of proof points, customer references, technology roadmaps and the latest enhancements to show how these vendors are on the leading edge of innovation. To reinforce these messages and position themselves in the broader context of how technology is evolving, each had high profile authors providing keynote addresses about their latest works.

At BroadSoft, Chris Anderson spoke about The Long Tail; his now famous work that explains how long-term value can be created and sustained in the Internet world. Sylantro countered with Geoffrey Moore, of Crossing the Chasm fame. He spoke more about his new book, Dealing with Darwin, and focused on how innovation is the key to survival in business. Both companies understand that customers expect more out their vendor solutions today, and these events are strategic platforms to show their vision about the transformational power of technology and how they fit into the bigger picture.

On a more practical level, there was a lot of common ground for both sponsors and exhibitors. Most notably, IBM, Polycom, Linksys, Siemens, CarrierAccess, CosmoCom, Covergence, Atreus Systems, Aastra, Acme Packet and Citel were sponsors of both events. Each event also had strong exhibitor support, with many vendors participating in both. Undoubtedly, this created some conflicts in deciding whom to send to each event, and the same held true for attendees. We had several conversations with prospects telling us they had to choose which one to attend and how they wished they were on separate dates instead of being together.

It appears that BroadSoft organized the bigger event, but to be fair, they do have a larger customer base, and perhaps are a more widely recognized brand name. Regardless, there appeared to be a lot of active networking at both events, and we received positive feedback from partners and customers alike who attended, stating that the conferences were better for executive-level networking than traditional industry shows like VON and Internet Telephony.

Both vendors are reaching critical mass now, and may have aspirations to follow in Acme Packet’s successful IPO footsteps. The roadmaps presented at these events show that both companies are maturing nicely, have rounded themselves out with strong management teams, established a rich ecosystem of technology partners and developers, and are well positioned for the emerging technology trends, namely IMS, FMC, SIP and mobility.

At Connections, BroadSoft did a great job displaying their new Call Center offering, and other desktop applications. On the mobility side, they announced BroadWorks Assistant, an FMC application providing PBX integration, which supports Nokia’s E Series phones, and they also showed some interesting demos with companies like FirstHand Technologies. There was also a lot of buzz among service providers on Polycom’s new IP650 wideband phone, which is compatible with BroadWorks, and how this could be a differentiating factor for hosted services over premise-based systems. Broadsoft also had some interesting future smart home-type services built around the “family butler”, who would serve as an IVR auto attendant for everyone in the home. The company’s software is now up to release 14, representing an unprecedented degree of maturity.

At the Sylantro event, there was a lot of focus on the importance of IMS, and Web 2.0-style applications, especially in support of Microsoft, one of their key strategic partners. Sylantro have recently announced a hosting deal with Microsoft for Exchange, Live Communication Server and SharePoint. Their latest release 4.1 of the Synergy platform can facilitate the handoff of calls between cellular and WiFi networks, and in partnership with Nokia, they demonstrated how release 4.1 enables mobile integration with the desktop. Continuing on the wireless front, they highlighted their partnership with Motorola to enable WiMAX nationwide for Wateen Telecom of Pakistan, which is reportedly the largest of its kind in the world.

We like what we see from both vendors, and this seems like a case of healthy competition that is driving the market forward nicely. BroadSoft may have a larger customer base, but Sylantro may have a slight edge on the partnership front with global partnerships with Microsoft, IBM and Motorola – though to be fair, Broadsoft also has relationships with both Ericsson and Lucent. We think both companies may be potential IPO candidates, and certainly both could be strong additions to the portfolios of the larger telecom vendors.

Among the other pure play application server vendors, NetCentrex (acquired by Comverse) may be considered in the same league as BroadSoft and Sylantro, though they have more presence outside North America. VocalData was another early player, but never kept pace, and was folded into Tekelec some time ago. The same can be said of LongBoard, but they have since re-focused on FMC and are no longer a factor in this sector. IP Unity is really a hybrid solution and has never been a first tier competitor.

Perhaps the greatest threat facing BroadSoft and Sylantro comes directly from the legacy vendors like Nortel and Alcatel-Lucent, as well as the larger nextgen vendors who have partial or no application server solutions, namely Sonus, AudioCodes, Cantata and MetaSwitch. At this stage of the game, we do not see the legacy vendors displacing the leaders with their own solutions, and they will continue to thrive as key partners who bring innovation, value and speed to market to both legacy vendors and the service providers.

4. Art of the Deal: Level 3 Acquires Broadwing

On October 17, Level 3 continued its consolidation play by acquiring fiber carrier Broadwing Communications for approximately $1.4 billion. This is a cash and stock deal where Level 3 will pay $8.18 cash and 1.3411 Level 3 shares for each Broadwing common share, equaling approximately $15.31 per share at the time of the announcment. Overall, Level 3 will spend $744 million in cash and issue 122 million shares for the acquisition. This may appear to be an expensive move, but Broadwing brings a lot to Level 3, including both wholesale and enterprise customers.

Level 3 has been busy in the past year, most notably acquiring WilTel, TelCove, Looking Glass, Progress Telecom and ICG Communications, as well as divesting Software Spectrum. The overall strategy is to keep pace with the incumbents as the demand for fiber continues to grow. This is probably not Level 3’s last acquisition, but Broadwing should position them nicely on the two main fronts of the market – wholesale and access. For that reason, the motivation behind this deal is more interesting than the mechanics and balance sheet impact.

On the wholesale side, where long haul traffic is its main activity, Level 3 has long been one of the world’s largest Internet backbone providers. With recent consolidation among carriers, there are now six majors serving the U.S. market – the three ILECs – AT&T, Verizon and Qwest – and the three major independents – Level 3, Global Crossing and XO Communications.

Aside from acquiring Broadwing so that others could not, this deal considerably boosts Level 3’s presence in metro access. While the majority of Level 3’s revenue comes from long haul, Broadwing’s revenues are about evenly split between long haul and metro. The net effect will be to double Level 3’s metro business, which provides local access to the enterprise market. This is important because long haul is a high volume, low margin business, whereas enterprise has higher margins, and an attractive upside, as bandwidth requirements will only increase with the adoption of multimedia IP services.

The deal adds about $900 million in annual revenues to Level 3’s base of $4.6 billion, and Level 3 expects the deal to contribute positive adjusted operating income before depreciation and amortization in 2007 and to generate $200 million to $250 million of such income in 2008. It expects the deal to contribute $200 million in free cash flow in 2009, including savings and total integration costs of $110 million to $130 million.

It is also worthwhile to note the quality of Broadwing’s fiber assets. They operate a 19,000-mile, nationwide optical fiber network, built on a Corvis all-optical backbone. Optical switching provides advanced capabilities such as real time and remote provisioning, but the Corvis network had been reaching its limitations in recent years given that the equipment provider no longer exists. Level 3 has already committed to moving to an Infinera backbone, and Broadwing recently agreed to use Infinera as well. It is likely that the majority of the Broadwing traffic will be migrated onto the Level 3 backbone, and the Corvis equipment will be decommissioned.

The Broadwing deal is the latest in a market consolidation among a rather tight-knit community. Level 3 previously acquired WilTel (formerly Williams Communications Group) which was Corvis’ other main optical network customer. Corvis acquired Broadwing to morph from an equipment company to a service provider, and later added Focal Communications for metro access. A year ago, numerous discussions were held among Level 3, Broadwing, and Global Crossing about merging together, but the talks eventually broke off. Most recently, Broadwing was in serious discussions with XO Communications, when Level 3 entered the picture once again, leading to the announced deal. Interestingly, the same day the Level 3 – Broadwing deal was announced, XO also made an announcement that it was no longer for sale.

This brings us to the future, and what Level 3 needs to do to survive. The demand for bandwidth makes this an attractive market, but scale and cost efficiency dictate that only the biggest and best operators will be competitive. Not only must Level 3 keep pace with the ILECs, but also large cable operators like Comcast and Time Warner, who are investing in their own network build outs. On top of that, all the conventional operators – both wireline and cable – must contend with perhaps the largest threat of them all – Google.

Speaking of Google, there has been speculation about them acquiring Level 3. This could be an attractive exit strategy for Level 3, but we see it as highly unlikely. The selling price would be too high, plus there is no sound reason for Google to take on all of Level 3's infrastructure, not to mention their debt. Google can acquire all the fiber they need elsewhere, and at a lower cost. Furthermore, Google has the most important asset of all – customers.

Looking ahead, this is going to be a very competitive marketplace, and if major bandwidth consumers build their own networks, there may be less demand for long haul carriers like Level 3. In addition, large Level 3 customers such as Yahoo! could be lured to other carriers who are able to offer a stronger value proposition, especially in terms of richer content deals.

Level 3 is no doubt aware of what lies ahead, and this scenario provides more context to explain the Broadwing acquisition. The bigger picture is that Level 3 must migrate away from the commoditized long haul business, and this move is a good step in that direction. We would only expect that any future acquisitions will be along these lines or even more focused access network routes, and anything that gets them closer to bandwidth-intensive content.

5. Financial Highlights