I just refreshed some market research data from Frost and Sullivan on the outlook for SIP trunks in the coming years. The data is for North America only, but it confirms the opportunity for edge devices that enable the service will be quite significant. F&S predicts that over 14 million SIP trunk channels will be provisioned to business customers by 2012 (up from under 2 million a year ago). Since this is primarily a cannibalization of the PSTN market, I like to convert this into T1 terms, and the quick math shows that translates into more than 600,000 T1 PRI circuits that will be replaced in North America alone. Frost and Sullivan indicates the corresponding service value is in excess of $9 billion annually.
T1 'replacement' is maybe not the correct description. Actually ‘reprovisioned and reconnected' is a probably a more accurate. SIP trunks still require plenty of bandwidth, in fact, using G.711 coders, you only get about 18 SIP channels over a T1 circuit provisioned for data, vs. the 23 DS0 channels delivered on that same physical link via a T1 provisioned for PRI. So many of the T1s currently provisioned for voice will need to be re-provisioned for IP and reconnected to the enterprise data network. And that reconnection will require a lot of new SIP aware edge devices to support that conversion. So does this ultimately become just another big Cisco win as the encumbant PSTN guys grudgingly convert their customer base to SIP trunking service, or is there room for some other edge device guys and competitive carriers to make their mark?