Home > Technology & Telecom, Unified Communications (IPT) > SIP Trunking: What’s Slowing Adoption?

SIP Trunking: What’s Slowing Adoption?

December 24, 2009 Leave a comment Go to comments

Why many businesses are still using PSTN PRIs and T1s, when SIP trunking purportedly can save up to 90% on telecom operating expenses?

The answers to this question are real and tangible, and should not be lightly disregarded as ignorance or intransigence. Some of the reasons for adoption not occurring at a lightning pace are technical. Other reasons are organizational or perception. For companies looking to sell SIP trunks or market products that ride along with SIP trunks, understanding these concerns can help craft better strategies and tactics that directly address – rather than avoid or dismiss – the underlying challenges that exist in the marketplace today.

How much can we REALLY save by using SIP trunks?

The savings that can be directly attributed to replacing traditional telephone company circuits with SIP trunking vary widely and are highly dependent on a variety of factors such as:
  • Location – Are the businesses located in the USA? Europe? Asia? Latin America? Africa? Costs for telephone company circuits, such as PRI or DSL, vary tremendously.
  • Number of locations – Is there a single location or multiple locations with PRIs or telephone company circuits?
  • Concurrent calls – SIP trunks are not like traditional telephone circuits where a circuit is needed for each call and, when all circuits are in use, others attempting to call in or out get a busy signal. With SIP trunking, the only limitation on number of users depends on the amount of Internet bandwidth available and the billing mechanisms of the SIP trunking provider. It is, therefore, very important to estimate the maximum number of concurrent calls required during peak periods so that a business only contracts for what they need.
  • The variable nature of the traffic – Calling patterns vary for most organizations. They burst to a high level during what telecom engineers refer to as the “busy hour,” where the majority of calls and minutes consumed take place. Knowing these patterns can help in designing a cost effective SIP trunking environment. Also, it is important to understand if calls – both inbound and outbound – are largely local, national, or international. In the world of traditional telephony, local outbound calls and calls into your telephone numbers were free. In some cases, SIP trunking is more like cell phone service where there is no such concept of “local.” Understanding the variability of your business traffic and how that impacts billing is an important step is assessing the financial impact of SIP trunking.
  • The volume of inbound and/or outbound traffic – Although some pundits are predicting that the cost per minute for telephone calls will go to zero in the not too distant future, that day is not yet here. SIP trunking providers have underlying costs for completing calls to or from the Public Switched Telephone Network (PSTN) and pass these costs on to their customers. Some calling patterns are largely inbound (think customer service centers). Some are largely outbound (e.g., telemarketing companies). Understanding these patterns will allow businesses to get the best quote possible from various SIP trunking providers.
  • Average length of call – This may not seem important to a normal business, but it is important to Internet Telephony Service Providers (ITSPs) who provide SIP trunks. There are many companies that use predictive dialers (Think of the calls you get when there seems to be no one on the phone when you answer – that’s a predictive dialer and the operator just answered someone else a fraction of a second before you picked up), or auto-dialers (e.g., robot calls). Most of these calls last less than 30 seconds. ITSP networks are often not designed for this type of traffic and, if your patterns are of this nature, the ITSP may not be willing to accept your traffic or terminate it once they become aware.
It is certainly not a given that any organization will save a huge percentage of their ongoing telecom operating expense without better understanding their specific requirements. The savings calculated for many businesses ranged from 20 to 70 percent. The savings in Europe can achieve cost reductions near 90 percent but have never seen anything close to that in North America. Nonetheless, even a 20 percent savings that goes directly to the bottom line is worth considerable consideration in today’s economy.

Change can be threatening 

Most voice services in medium to larger enterprises are the responsibility of the Telecommunications group. The telecom manager understands voice, telcos, PBX systems, and so forth inside out. SIP trunking is a data service that runs over the IP connections that are generally managed by the Information Systems group. Telecom managers may be hesitant to relinquish control of a large part of their operating domain due to concern about the erosion of their positions and change in an organizational reporting structure. My personal view is that telecom managers still have a very important role to play and need to evolve as telecom evolves. However, this could easily be one of the underlying reasons why corporate telecom managers would not leap to a SIP trunking solution.

The status quo is not threatening

PSTN services are proven technology. They are easy to understand and, for the most part, reliable. They do not require addressing NAT or firewall issues (again the domain of the IS team). The PSTN may be more costly and relatively inflexible, but it is a known commodity. IP telephony can be impaired in a variety of ways, including the following:
  • Bandwidth contention – SIP trunking uses Internet connectivity to reach the outside world. Depending on the codec (compression/decompression) algorithm being used, each voice call likely consumes 50 to 80 kbps. Voice is extremely sensitive to latency or delay and we hear this as drop-outs from words or other interference. If the total bandwidth available at a particular location is reaching saturation – even for a second – we hear that disruption in a VoIP call. Someone downloading a large data file, video file, or other bursting applications (e.g., Multiple Outlook instances executing a send and receive within milliseconds of each other) can easily capture a large percentage of the available bandwidth. It is for this reason that bandwidth utilization must be carefully considered during the initial network design and constantly monitored to ensure that sufficient capacity is available.
  • Lack of appropriate traffic shaping tools – Because voice over Internet uses packets of information rather than a dedicated stream using a circuit, these packets can be identified and prioritized. Traffic shaping tools allow voice packets to receive higher priority than e-mail or file downloads and can thus provide improved quality of service and the ability to manage around unexpected bursts in traffic load.
  • NAT and firewall interference – Network Address Translation (NAT) is used to protect corporate networks by only allowing one Internet address to appear publicly external to the LAN. This helps to prevent attackers from reaching sensitive servers and proprietary information. At the same time, however, this also makes it difficult for voice traffic to navigate through an organization’s perimeter defenses and establish a high quality session for voice services. NATs and firewalls are wonderful components of an information system security plan, but voice was not previously part of that consideration. There are solutions to this challenge including a new breed of SIP-aware firewalls and appliances that are designed to address these issues, but that means adding or modifying elements in the network security plan. It may be inevitable, but change from the status quo is not an easy decision.
VoIP has security vulnerabilities
Internet telephony is an application that functions via Internet has the potential for security vulnerabilities just as networking in general. PRIs are perceived as secure and certainly do not threaten to serve as an attack vector for potential attacks as Internet access may. Clearly, SIP trunking presents vulnerabilities that need to be directly addressed including SPAM over IP Telephony (SPIT), the potential for Denial of Service (DoS) attacks using IP telephony, theft of SIP signalling, eavesdropping on communications, and many others. All of these can be addressed successfully, but that takes knowledge and effort that may be scary to potential customers or executive decision makers, and unfamiliar to telecom managers.

I have worked with a number of companies on these concerns, but it is an evolution rather than a sudden change of mindset. Business people still have legitimate issues that can be answered with proof of concept trials and the critical mass of others that have successfully moved in this direction. SIP Trunking can be a phenomenal solution and will be adopted by a growing number of businesses worldwide when their justifiable trepidations are answered professionally.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: