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. Skype Outage – Down But Not Out

eBay’s landmark investment in Skype has had mixed reviews from the beginning, and the service outage that began on August 16 has led only to more questions, most of which cast doubt on what they actually got for their money.

On August 16, the Skype network suddenly went down around midnight, Pacific Daylight Time. With fits and starts, service was fully restored in a couple of days, and has remained in seamless operation since. IP networks are generally not expected to stay up 100% of the time, and what caught widespread global attention was the fact that the Skype blackout happened so suddenly and so quickly – with the whole network being disabled. This has never happened before, creating great cause for concern, not just within the Skype community, but at parent company eBay as well.

Skype has built a tremendous and loyal following, with some 200 million “unique” registered users globally. In absolute terms this is massive for any type of operator, and is a key reason why Skype was acquired by eBay. Being a largely free service, the meaning of that metric is rather dubious, but at any given time, there are typically seven to nine million people using Skype worldwide. This is a more realistic indicator of Skype’s actual usage, and it is still a substantial amount of activity. However, most of the traffic does not generate revenue, and for this reason their users cannot really be called subscribers or customers.

Regardless, Skype’s outage impacted a large number of people globally. Whether they rely on Skype for instant messaging or VoIP, millions of users had to make do without Skype for a few days. Since there is no monthly subscription charge for Skype, the revenue impact was minimal. That may be the only upside here, as we see several potential issues that present some important and new challenges for Skype to address.

Although Skype has made steady progress in building up its revenues and evolving a viable business model, ARPU is very low, and we suspect that the vast majority of users only use the free features of Skype. As such, when a free service has problems, there may be little financial fallout, but the loss of confidence and trust can be far more costly. Another key reason why Skype was acquired was the extremely low cost of acquiring customers/users. So long as Skype can continue its staggering growth in adding users at virtually no cost, the potential for profitability remains attractive.

That equation may have become jeopardized with the service outage, especially since there are still many unanswered questions. For Skype to become a good investment for eBay, it must continue acquiring users, and to do that, the service must be reliable and it must scale without having outages like this.

At the heart of Skype’s brand is trust, and when this many people become inconvenienced by an outage, they are entitled to a good explanation and a reassuring resolution. By most accounts, Skype has not delivered on these questions, although service has been fully restored. Skype is a media and tech-savvy company, and their low profile response was surprising. Initially, it appeared the outage was caused by Microsoft’s “patch Tuesday”, which Skype was slow to refute, giving credence to this explanation. However, there was nothing to suggest this particular patch update was different from previous updates, and furthermore, Mac Skype users were also affected. Eventually, Skype came around to acknowledging this was a software problem at their end, and have since accepted full responsibility for the outage.

Skype’s slowness to provide a clean explanation for an unprecedented problem has given rise to seeds of doubt about their technology and the reliability of peer-to-peer for large scale use. It also raises questions about Skype’s ability to anticipate these problems and make their network robust enough for the growth that is expected – and required – for long term success.

Most of the attention on their technology has focused on Skype’s proprietary architecture for peer-to-peer, especially the use of supernodes. Essentially, Skype’s model is client-based as opposed to being server-based. The use of supernodes allows Skype to scale with minimal cost since the computing power is distributed in a decentralized fashion among the Skype users themselves – hence, being client-based. The trade-off for low cost scaling is a lack of control over the network, which is precisely the reason why most operators use a server-based architecture. This model is certainly more expensive, but it provides far more control for the operator, and we concur with the view that Skype may not have gone down if this was their architecture.

In this regard, there is some concern that the Skype outage will cause the market to further question the suitability of peer-to-peer for applications such as VoIP. We feel this would be a mistake, as other peer-to-peer models exist, and could well gain critical favour, especially for the enterprise market, which has largely been off-limits for Skype. One alternative in particular is SightSpeed, a company that we have cited in the past. The key difference in their approach to peer-to-peer is the use of a standards-based and SIP-based architecture. They do not use supernodes, and rely instead on a server-based model, which allows them to ensure a high quality end user experience across all types of network conditions. Furthermore, by conforming to industry standards, they can support a wider range of technologies commonly used by enterprises, and more importantly do not run into the firewall blocking issues that prevent Skype from being used in a meaningful way by enterprises.

2. Fall Industry Conference Preview

September marks the resumption of telecom conferences, and given the amount of change over the past few months, the Fall shows will be a telling barometer of what we can expect going into 2008. Just considering our newsletter, we have recently seen large and small acquisitions, strategic alliances, privatizations, and both successful and unsuccessful IPOs.

Change is a constant, but consolidation has been the dominant trend. This may be good news for the bankers and VCs, but a shrinking pool of both carriers and vendors can be problematic elsewhere. The competitive environment is getting much tougher for Tier 3 players, who now have fewer customers to sell to, and narrower options for selling into larger customers, where the long-term returns are much more attractive. Things are even harder for start-ups who are looking to enter the market, and are one level below the Tier 3s in the food chain.

We note this because this is where so much of the innovation and disruption comes from. As the consolidators come to dominate the market, there is a valid concern that these forces will lose momentum. IP communications is on its way to becoming mature and mainstream, but the innovation curve is far from complete. It is our hope that the market will continue to support a diverse ecosystem that includes new entries and companies with creative visions.

This is also a valid concern for the conference producers, who need a diverse pool of supporters to keep their shows going, both as attendees and exhibitors. The upcoming TMC Internet Telephony Expo is a good example of this, and is the first major Fall event, running from September 10-12 in Los Angeles. Even a show that is this well established needs to position itself today as an essential event, and is billed as “The World’s Ultimate IP Communications Conference”. To deliver on this promise, TMC has an ambitious agenda, addressing a wide range of IP-related topics – FMC, IPTV, IMS, SIP, Open Source, Unified Communications, Security, Peering, WiFi, etc.

There is certainly something for everybody, and in this regard, the conference will likely provide great value for attendees. IP has become so pervasive that it becomes necessary to take this approach and ensure that all the main touch points are covered. Conferences are typically either horizontal or vertical, and TMC is a great example of the former. They will, and do, attract a certain audience, and provide a good base of content for attendees to build on. However, there is also a strong market for vertical conferences that focus on a specific sector – such as SMB or an industry such as financial services – or a technology space such as VoIP, IPTV, IMS, etc. These shows are more in-depth and focus on both emerging trends and business issues. They tend to be smaller scale or more regional than horizontal shows, but are attracting a growing following. This is particularly true in IP communications, where the market has generally moved beyond the early adopter stage, and attendees are looking for a more substantive experience.

We see this posing a challenge for shows like ITExpo, VoiceCon, and VON, where the entry level pool of people seeking a broad base of content is getting smaller. For people who have regularly attended these shows for some time, their needs are changing and often becoming more focused around a particular technology or market. This may well be driving them to more specialized shows, and as this trend continues, exhibitors will follow. As such, there may be too many shows chasing too few people. Conferences are expensive to produce, and expensive to attend, and our sense is that the IP communications space has become saturated with events.

In our view, this is symptomatic of the issues facing vendors and carriers. With consolidation whittling down their numbers, there are fewer of them able to support conferences, which also means they must be more selective about which shows to invest in. This suggests that consolidation may need to come again to the conference market itself, which has gone through this cycle several times before. We all remember the Supercomm, Globalcomm, NXTcomm split/merger debacle.

Pulvermedia’s VON franchise is the other major show in this market, and in June, they raised $11 million from Technology Investment Capital, likely in anticipation of just such another cycle. We would not at all be surprised to see some major movements among conference producers next year, as they all struggle to adjust to the consolidation trend among their customers. In that regard, we think that September’s TMC ITExpo in Los Angeles and October’s Fall VON in Boston will provide a preview of what 2008 holds in store not only for the IP communications sector, but perhaps for the conference producers as well.

3. Cisco and Microsoft – Too Close For Comfort?

On August 20, the two biggest companies in tech made some very public pronouncements about their intentions to work more closely together. Microsoft and Cisco appear willing to practice what they preach in terms of the big vision they are both selling to the enterprise market – collaboration. Just as they are focusing heavily on solutions that make it easier for people and businesses to collaborate, these two companies are planning to do the same across seven areas, including security, mobility, IT architecture and unified communications.

The two companies are intense competitors, and in particular, seem to be on a collision course in the emerging area of unified communications. They have fundamentally different views here, where Cisco believes that the network is the logical driver for integrating all forms of enterprise communication. Conversely, Microsoft sees the operating system and software as the ultimate platform to drive unified communications. Both are betting heavily on their visions of unified communications, and while both solutions are still works in progress, each has merit and will find a healthy market in time. However, the stakes are very high in that the lead vendor will essentially own the customer, and it is difficult to see how they will truly co-exist in shared accounts.

That said, their announcements indicate there are several other areas where they can effectively collaborate. Intuitively, there is no pressing need to do this, but the stated rationale is that their customers are demanding it. This sounds simplistic, but it reflects some difficult realities these two companies are facing. Both are dominant in their respective lines of business, and as they expand into new areas to find new sources of growth, they are increasingly stepping on each other’s toes. This is creating difficulties among IT decision makers who often must choose between them because of their limited interoperability. The implication is that these companies had better learn to work together, or both risk losing business to competitors who are easy to interoperate with.

Essentially, both companies recognize that a monopoly position is not possible, but a duopoly is, providing they collaborate in the right ways. “Co-optition” is the best way to describe this relationship, and to protect their customer base, Microsoft and Cisco will collaborate not just on the product level, but at the human level. Sales, marketing and even executive teams from these companies will work more closely together to create a consistent vision and message so customers understand how to work with them.

This may be a customer-friendly message, but it raises questions for everyone else. With most of the other enterprise vendors having made recent consolidation moves, this partnership may prove challenging, and will put added pressure on competitors to find new ways to live with the new status quo. This is particularly true for Nortel, who already has a partnership arrangement with Microsoft for unified communications. Since it is not exclusive, no agreements have likely been breached, but this may not be good news for Nortel. It is too soon to tell if this reflects a lack of confidence in Nortel as a leading partner with Microsoft, but nonetheless is a welcome development for Cisco.

Finally, we need to look at how this move benefits the two partners themselves. Cisco excels at the IT level, but it remains to be seen how well they can sell platforms and solutions that directly impact the end user. Microsoft can certainly help on that front, but to what extent is not clear. Cisco has also learned how to achieve market dominance without attracting the negative attention that hurt Microsoft in the past. Furthermore, Cisco is more adept at succeeding in new markets, whereas Microsoft has had mixed results. This was reflected in Cisco’s recent Q4 earnings, where the news was positive, and the company can feel confident entering this alliance from a position of strength. Microsoft may have the upper hand at the desktop – the last frontier for Cisco – but we see Cisco standing to gain more from this relationship if the two can truly work together.

We should add that despite the dominance of these companies, there may well be a common enemy they also need to be concerned about – Google. On many levels, the threat posed by Google is well understood by these companies, but they are equally concerned about the potential for Google to move in directions they have not yet seen. Joining forces may well be as much about keeping their established competitors at bay as much as ensuring that Google does not become established in this market. To a lesser extent, this thinking also holds for Apple in the consumer world, but in the enterprise communications market, we see this collaboration being very much driven as a means to keeping this a two-horse race.

4. Art of the Deal: NEC Acquires Sphere Communications

In a departure from the string of large scale acquisitions that have become commonplace in 2007, this month’s deal story is very modest, but still notable. On August 15, NEC of Japan acquired privately-held Sphere Communications, which is based the Chicago suburb of Lincolnshire, IL. NEC is certainly large enough to have made a major acquisition, but the $42 million they spent on Sphere is an interesting choice.

At face value, the appeal of this deal is easy to see. NEC is a major PBX vendor, transitioning from TDM to IP, and is known more outside the U.S. in the business telephony market. Sphere is a much smaller vendor, but highly focused on enterprise IP communications, and selling mostly within North America. The business telephony vendor space has been undergoing unprecedented consolidation recently, and in this regard, the deal is good news for Sphere. For NEC, though, we see this deal as more of a technology acquisition than industry consolidation. Sphere does not bring them much critical mass, and there is limited overlap between product lines or customers. As such, this is more of an integration deal, where NEC sees Sphere as adding significant value to their IP telephony offerings.

Sphere has a relatively small customer base that spans both ends of the market. They serve some large enterprise and government customers with their Sphericall IP PBX, though many of their deployments would fall into the SMB category. Being IP-based, they also serve a third market – developers and ISVs (independent software vendors) – with their Communications Services Engine. As a small company, Sphere also has a fairly small channel partner network, though one public partner of note is Polycom, which packages its IP telephones with Sphere's software in at least one configuration.

As a pure software product, Sphere’s scalability is very good and can serve deployments up to 30,000 seats. As such, Sphere has the capability of enhancing existing PBX and IP PBX systems, but also as a developer and platform for new software and applications that leverage IP – SIP, SOA (service oriented architecture) and Web services. Formally, Sphere will become a subsidiary within NEC’s Unified Communications Software Development division. It is likely hoped that NEC's much larger sales and marketing channels will give Sphere the chance to become a more relevant player in the IP PBX market, both in North America and globally.

Turning back to NEC, their motives also appear to be survival-based. In North America, they are ranked fourth in the PBX market, behind the Big Three – Cisco, Avaya and Nortel. No hardware vendor on the horizon can truly challenge Cisco today, Avaya has recently gone private, and Nortel appears to be hinging its hopes on an alliance with Microsoft. There are no logical Tier 1 vendors for NEC to partner with, so to make a move in this market, acquiring a company like Sphere is the next best thing. With the recent Mitel/Inter-Tel merger, they are also a player in this market, though the majority of their customers fall into the SMB category.

Interestingly, the acquisition was very low profile, especially in North America. There was an initial press release on the NEC website in Japan, followed a week later by press release on Sphere's own website. We suspect this was intentional, although it could be argued that the small size of the deal alone would have ensured limited visibility. It is more likely that NEC did not want to draw much attention to the deal, as it signals to others their intentions in the IP PBX market. Not only does Sphere’s technology give them a stronger foothold on the ground, but it considerably strengthens their IP transition story for their customers. The faster they can integrate Sphere’s applications and interfaces into their IP PBXs, the faster they can challenge the big three in this market.

It is also interesting to note that both Marconi (now Ericsson) and Allied Telesyn, also from Japan, were previous investors in Sphere. Presumably both companies also had an opportunity to buy the company, but chose not to do so. Sphere dates its beginning back to 1994, so it has remained an independent company for a long time. Perhaps it was just a timing issue, but perhaps NEC was able to see some potential in the company that the others were not.

Aside from the traditional competitors, we think there is an even bigger reason driving this deal. It is none other than Microsoft, who is working very hard to become a complete business solution that will soon include voice, especially under the Unified Communications umbrella as well as OCS 2007 – Office Communications Server. They are not there yet, and today cannot seriously render the PBX vendors obsolete. However, the potential is there if they can deliver on their promises and ambitions, and that is making all these vendors nervous.

As such, there is a time-to-market issue facing all the PBX vendors, as they need to protect their customer bases from the Microsoft machine before they really comes to market, and to entrench themselves with new customers before they turn to Microsoft for everything. In this regard, software-based vendors like Sphere have great value, as software solutions evolve and adapt more quickly than hardware, especially against Microsoft, who is the ultimate software vendor.

5. Financial Highlights