Enterprise UC

Unified Communications in the enterprise is a huge 13-15B business as of 1Q 2011.  Market ownership is divided almost as a duopoly between Avaya and Cisco.  There are several reasons why smaller players like the ones noted in the research below are unable to crack this duopoly.  Some of the more important ones are as below:

  1. Large companies like Microsoft are solely software vendors – they try to build UC solutions( Lync, Exchange )  with OEM vendors (AASTRA) providing phones and gateways. This never works because inter-opping a piece meal solutions is easier said than done.  Enterprises are not  interop forums – they need a communications network right off the shelf.  Lack of hardware infrastructure (both manufacturing expertise and also internal leadership experience and strength) is an insurmountable weakness.  Defensive acquisitions can give these companies that much needed hardware suite to go with their strong software offerings.
  2. Siemens, IBM and Alcatel-Lucent are legacy and old world conservative companies with monolithic solutions – transforming those humongous PBX “hall size hardware”  into pure IP solutions  for the enterprise market is more than a little challenging.   What we have now are patchwork TDM solutions with IP side functionalities = these firms implement their legacy feature sets onto these hybrid solutions and hence gain a sliver of the segment.
  3. Most of the larger PBX  based firms start at the top of the product  hierarchy working their way down with contact center solutions, SBC’s and high end gateways which are hacked versions of their  legacy PBX’es.   These still have interfaces and API’s which make it difficult to support current applications like CRM, Salesforce.com etc.  Another way of stating this is that services and features sell solutions and not the other way around.
  4. Smaller firms like Interactive Intelligence and Mitel try to swim up the river by focusing on scalability – this comes a cropper as well because of the lack of intrinsic scalability or 5 or 6 9’s in their fundamental design.  This is like trying to load up the Prius with more power so that it can compete with the muscle cars at Nascar  – a very “average” idea indeed.
  5. Somewhere in the middle lie Cisco and Avaya. Cisco with its strong routing and switch background easily implement and support large enterprises  with IP solutions and so enterprise UC is merely an extension of their DNA. Avaya built an enterprise based company with a specific segment focus and hence they rule as well.

Cloud VoIP

Most of the “cloud basedVoIP solutions involve putting the software running the switches and gateways on the cloud.  For now lets not confuse these services with cloud based voip calling services.  Here our focus is on cloud based PBX solutions.  These are basic call handling software appliances which are capable of routing calls into different endpoints (which could be either physical desktop sets or web based call managers  or mobile handsets) .  The “cloud” part of the solution involve several aspects but the most important ones are :

1. Access to large resources like bandwidth, hard disk and memory ( real time traffic like voice cannot have latency and/or delays )

2.  Security for signaling traffic

3. Redundancy and high availability

4.  Scalability for customers who want to grow their installations ( or remove idle machine usage during holidays etc )

Key features expected in a IP-PBX solution :

Lets assume we use a combination of Amazon EC2 and Asterisk to implement a cloud based VoIP solution.    Amazon has solved the persistent data issue and IP address loss issues with EBS and Elastic IP addresses respectively. Both of these are vital to the running of a VoIP ‘instance’ because it needs storage for voice mails, logs, DB etc.  Also IP addresses cannot be dynamic since everytime the instance was shutdown, the next iteration would yield new hostnames and IP addresses.   Asterisk is chosen  for its open source ability to be hacked up on a server.  Here is a quick cost breakdown for such an installation.  Phones are needed to make this configuration work as opposed to a pure desktop call manager topology.


Related Articles

Total Cost to own a complete UC solution

Let’s take a look at the commonly used unified communication implementations. Here are the usual pieces of this puzzle:

  1. Phones and switches ( mostly full IP based meaning old world TDM solutions are phased out )
  2. Gateways with PSTN connectivity ( T1, E1, FXO, FXS, E&M, SIP Trunking )
  3. Collaboration software ( like WebEx, Citrix solutions etc )
  4. Email solutions ( Outlook, Zimbra et al )
  5. Messaging and presence systems ( Agito, DiVitas etc )
  6. Bridge conferencing – large scale audio and video conferencing ( telepresence also falls into this category )

Several strong competitors in this market (Cisco, Microsoft, and Avaya) make this segment very attractive to follow. Cisco with its strong strengths in #1, #2 and #6 seem like a slam-dunk solution but Microsoft with its ubiquitous position in the #4 and #5 is also positioned for success. Success in this segment is measured across several smaller sub-sections which are the SMB, enterprise and service provider segments. For now let’s focus only on the enterprise part of the market – in this context we will look at a firm with 1000 seats. I have chosen different vendors for each of the solutions to give a good idea of the cost incurred – this knows that no one vendor offers everything in a bundle though one or two claim to do so.

Summing, it takes a good million and change to own a Unified Communication system. Now we have not included the training for maintenance, the integration costs for each of the components and the recruitment costs for IT personnel. All of this may easily come to another million. A 1000 user company is a full fledged enterprise with very high UC costs ( doh!). So why don’t we move all of those costs to the cloud.

In the next post, I will detail all of the costs involved in moving each of the above components to the cloud and how it will affect operational costs.

Gartner and UC

Someone asked me recently about defining UC and this post immediately came to my mind : http://blogs.gartner.com/nick_jones/2010/09/07/is-unified-communications-the-biggest-scam-since-ponzi/

If a senior analyst at Gartner is skeptic  about the mish-mash of solutions all bundled together ( aka UC ) , I wonder how customers truly feel when they have opaque solutions from different vendors on their desktop.  It possibly isn’t too different from that of the analyst albeit mostly usability reasons !

UCIF update

More vendors are gravitating towards the consortium.  VXI  ( http://www.marketwatch.com/story/vxi-corporation-joins-unified-communications-interoperability-forum-ucif-2010-08-27?reflink=MW_news_stmp ) joins the bandwagon and I must admit they look pretty strong now.   Now waiting for other biggies to join.  I am more interested in the interop and standards that these firms collaborate in this space as opposed to management-speak for a team ( aka consortium). We had SIG’s and other technical standards committees driving some of the developments for a while and now its seems at a higher level.   UC interop is a huge market and I think of all the firms it would probably be the desktop and the network that will win – Microsoft and Cisco/Juniper.   More to come soon…