Can’t We Just Dump the PBX and Go All-Cellular?
Can’t We Just Dump the PBX and Go All-Cellular? by Michael F. Finneran
As a greater percentage of workers become mobile and start to use their cell phones as much or even more than their desk phones, inevitably the option of going “all-cellular” comes up. Indeed workers, and particularly the more mobile workers, are already making and receiving the majority of their calls on mobile phones, however for business users, there is still a lot you give up when you go from a PBX to a mobile network.
The first thing to recognize is that cell phones are essentially “personal communications devices.” From a voice standpoint that means the operators are looking to provide the same range of services you’d get on your home phone; there is no concept of the user being part of a larger “organization.” That “personal communications” concept is why when cell phones first appeared, many people would give you their wired number, but they’d give their mobile number only to people they really wanted to talk to. Even today most people are more guarded about who they give the cellular number to. The fact that U.S. customers pay for inbound calls probably contributes to that as well.
There is no denying the fact that mobile phones, particularly smartphones, are more convenient to use with their integrated address books and the ability to dial a phone number from a calendar entry, a document, or a web page; UC solutions are starting to bring some of those capabilities to the desktop, both in the UC client and in the phone itself.
However, there are still a lot of business calling features that cell phones lack, including:
- Attendant or Auto Attendant support: There’s no “main list number” with a cell phone, it’s all “DID”
- Call transfer
- 6-party conference
- Do Not Disturb: You can turn the thing off, but you’ll probably forget to turn it back on
- Hunt groups
- Group pick-up
- Paging access
- Abbreviated dialing: This is less important given the integrated directory
- Boss-secretary capability
- Contact center functionality: Though it would be hard to imagine anyone running a contact center with cell phones
- Private networking to reduce cost
For years people have talked about adding PBX-like features like these to a cellular service, generally using the term “Mobile Centrex,” however the mobile operators see their market as a “consumer market,” and there appears to be little interest in spending a lot of money on stuff that would have value only to business customers.
Besides the limited business telephony feature set, cell phones face other challenges in serving as a primary business phone line. First off, they run on batteries, and business calls need to get through. That fact is obvious to business users who are inevitably bunched up around the charging stations at the airport and are looking for the seat next to the electric outlet at Starbuck’s.
Also, the sound quality is often terrible. We give up a lot to be mobile, and sound quality is at the top of that list. Some years ago I mentioned the quality problem on cellular in a talk I was giving and observed that I had never seen anyone quoting mean opinion scores (MOS) for cellular networks. The next day one of the engineers in the group brought in what appeared to be an internal paper by two engineers from Cingular Wireless. The two had done extensive testing on the Adaptive Multi-Rate codecs used on GSM cellular networks.
The paper contained mountains of data (these guys were “thorough” with a capital “T”) and compared the performance of both half-rate and full-rate codecs operating at data rates between 4.75 and 12.2 Kbps in a variety of network conditions. The long and the short of it was that there were almost no samples exceeding a MOS score of 4.0 (typically the minimum quality level we’re looking for in enterprise telephony) and plenty below 3.0 (generally the level associated with “Get your resume in order!”).
For a routine contact, the quality is only moderately annoying, but we regularly have to mute callers who have dialed into the conference bridge on a cellular phone. Also, if you’re doing a webinar, the engineer will always tell you, “No cell phones!” Of course the codecs that were used in 2G networks like GSM were chosen based on the operators’ need to minimize the bandwidth consumed on a network with limited capacity. We’re waiting to see if and by how much the quality will improve as carriers shift to voice over LTE (VoLTE), which will be able to support broadband as well as traditional narrowband codecs. That transition is due to begin later this year.
In the end, whether or not “all-cellular” is an option gets down to how your users communicate. For the most part, in large enterprises everyone still gets a business number – even if they spend only a small part of their time in the office. And that number will typically ring the user’s mobile number simultaneously when it’s called. Many small businesses have essentially gone all cellular, and if you call the office number it goes straight to voicemail.
However as large businesses come to grips with how (and “where”) their people are working today, we’ll come to see far more use of hoteling, and while every user may have a business number, it may only be associated with a desk set when the user logs in at a hoteling facility.
So I can’t foresee the complete disappearance of the PBX (or “UC platform”) in the near term, especially for applications like contact center, though mobile workers are moving en masse to mobile phones and home office phones. The mobile operators have essentially no interest in developing any mobile voice service specifically targeted at the business user, we’ll have to look elsewhere for a means to better support those mobile users. Sure, simultaneous ring is a start, but the IP PBX/UC&C haven’t been able to get beyond that. Almost no one uses the mobile UC clients they offer that would allow a user to transfer a mobile call to a desk phone (assuming the user even has a desk phone – or a “desk” for that matter), so none of those mobile users is getting any value from that seat license we’re buying. Maybe what we should be looking for is a session manager that knows our DID numbers (some of which are assigned to mobile users), and if it sees a call to one, it simply forwards it to that user’s mobile number.
So while we are moving toward the “Consumerization of IT,” it appears that the transition to the “Consumerization of Enterprise Telephony” has already occurred.
Note: I’d like to thank my good friend Marty Parker for his review and input on this piece.