A number of articles on NoJitter.com have been emphasizing savings from shifting voice communications to SIP trunks, including the top notch "SIP Trunking Best Practices" by Sorrell Slaymaker, "How Much Does SIP Trunking Really Save You?" by Eric Krapf, and the mention in "A New Public Network?" by Fred Knight. These savings are another phase of least cost routing, where enterprises invest in premise equipment and software in order to cut their recurring carrier billings.
However, the articles do not point out two important factors:
1. Voice traffic over carrier trunks is on the decline. The number of voice minutes per user is being constantly reduced by the newer forms of Unified Communications (more below).
2. SIP Trunks are a method of connecting voice traffic to the Public Switched Telephone network, not a platform for Unified Communications. UC systems send voice and other media (both peer-to-peer and conferencing) over the private enterprise LAN and WAN data networks and over the Internet. The UC systems may use SIP protocols, but only use SIP trunks for residual PSTN traffic to parties that are not on or federated with the UC systems.
For those reasons, most enterprises will benefit by first preparing a Unified Communications and Collaboration roadmap. That UC roadmap can then be the basis for updating the forecasted (likely much lower) requirements for voice PSTN trunks before committing to a move to SIP trunks. The justification of this approach is that the precious capital and Telecom/IT staff time can be spent on the highest ROI activities. Usually, this will be the UC investments which can reduce costs in many parts of the business, not just in telecom transmission billings. If an enterprise can afford to do both SIP trunks and UC, the planning is still useful for proper sizing of SIP trunk requirements and investments.
Here are some highlights of how voice communications traffic is being reduced or is being converted to IP data network traffic by Unified Communications and Collaboration solutions.
* Instant Messaging (IM) and Presence have been shown to reduce voice traffic dramatically in the enterprise. Federating to public services and to other enterprises can extend these traffic reductions to communications with business partners and with some customers.
* Communication-enabled business processes (CEBP) via both mobile applications and application portals are shifting communications to data interactions or to IP voice on the data networks.
* As video replaces voice, that video traffic will leverage expanded data networks. Video communications are either peer-to-peer IP communications between the video end-points or IP connections to video conferencing servers.
* Mobile voice communications on smart phones and tablets using emerging VoIP and video apps over the cellular and Wi-Fi data networks will convert another chunk of voice traffic to secure IP data channel traffic, bypassing the PSTN trunk connections.
* Social networks and collaborative workspaces (for businesses) such as IBM Lotus Connections and Quickr, Microsoft SharePoint, Skype for Business, Google Docs, and Cisco Quad also replace some amount of voice traffic; remaining voice traffic is via the UC methods suggested above.
These trends are already quite visible in case studies and other reports. First, voice usage is being replaced by other tools. Second, UC systems manage voice over the enterprise data networks and the Internet, using SIP or TDM trunks only when necessary to connect calls through the PSTN.
Note that moving voice traffic to the LAN/WAN and Internet is even more cost efficient than converting to SIP trunks. Both the transport and transit of secure voice transmission over the LAN/WAN and Internet are typically less costly than even SIP trunks. A savvy midsized enterprise can source North American WAN services for less than US$10 per megabit per month (the equivalent of about 8 VoIP sessions using G.711 or about 16 VoIP sessions using G.729 with max bandwidth usage at 60% of capacity).
Further, none of the new UC&C communications modes listed above require Session Border Controllers (SBCs) except for the residual SIP trunk connections to the PSTN. There have been some posts suggesting that SIP Trunks with SBCs establish an end-to-end IP connection which would provide the route for UC communications such as those listed above. Sure, some SBCs can also provide Internet firewall and threat management services, but most enterprises already have a deployed standard (vendor, protocols, and methods) for Internet protection and security. The new UC communications will more likely be routed through that existing infrastructure capacity rather than over equipment from new vendors requiring additional investments. Convincing a CISO (Chief Information Security Officer) to "sign off" on having a nouveau voice platform displace his entrenched and regulatory-compliant UTM (Unified Threat Management) platform, is likely a losing (or very long lead-time) proposition.
So, as we have been emphasizing for some time at Enterprise Connect (nee VoiceCon) and Interop and at UCStrategies.com, it is really important to prepare a UC strategic plan and the resulting UC investment roadmap before making new investments which just migrate the old PBX approach (phones, switches, cabinets, trunks, adjunct application servers, etc.) to new technology platforms (IP phones, communication servers, session managers/SBCs, SIP trunks, adjunct (virtual) application servers).
If you want the biggest savings, change how you do business, rather than only changing the tools for doing business the same old way.
