Table of Contents:

i. Background
1. Inability to perform overhead paging using Cisco SCCP phones.
2. Insecurity of Win32 platform on main Cisco softPBX imposes great overhead.
3. Meet-me paging applications are primitive.
4. Cisco's IP phones are too expensive.
5. Cisco's E911 responder servers add risk to a critical aspect of telephony.
6. The exclusively-distributed approach to telephony switching adds unnecessary failure points.
7. There's no program for 24x7 system monitoring provided by Cisco.
8. There's a hug feature gap between CallManager and CallManager Express, making large system design more difficult.
9. Cisco's legacy of non-support for 802.3af is hurting its customers in the long-term.
10. SIP endpoints can't be supported by the CallManager, making Cisco's softPBX a poor choice for service providers.
ii. Conclusion and Recommendations

Between 2002 and 2004, I embarked on an ambitious research project, the object of which was the total replacement of a large multi-company enterprise PBX network with a centralized, stable, and feature-complete VoIP network. My biggest objective was the replication of features that existed in our legacy PBX network-an aspiration that I later discovered was a bit misguided. The VoIP technology family is replete with features and nuances that make over-arching consideration of “the way things are today” a wasteful endeavor.

Coming from a data communications background, my first real encounter with Voice over IP was Cisco's AVVID/CallManager environment. At the time I started the project, this platform was the clear market leader for IP-based call-switching, having shipped more than double the number of IP-based phones than any of their closest competitors.

I myself was a sold-out believer in Cisco's IP Telephony system, having received some shiny, well-done demonstrations from a Cisco VAR. But by the time I finished my research, they had slipped to number three in the market for two consecutive quarters behind Nortel and Avaya. While their position as a realative newcome in the telephony world has hurt Cisco recently, I don't excuse their market share slip on the basis of that fact alone.

Instead, I decided to find out what is wrong with CallManager that caused the slide to occur. I believe that when implementers of this technology decide on a vendor, they are really deciding on a set of features and benefits that drive costs and affect operations. For this reason, I found myself more and more on the opposing side of Cisco's sales pitch. As much as I resisted the notion of not choosing Cisco, if merely because of a warm affinity for Cisco as a company, I ended up quite opposed to their approach to telephony.

Many of the shortcomings of Cisco's VoIP platform are so esoteric that they can be overlooked, but possibly at the expense of stability or telephony features. I didn't want to run these risks on my network, so I forced the issues with Cisco. To their credit, I did get reasonable responses to most of the issues I brought up. The VAR I was working with brought in four different groups of Cisco reps and engineers over the course of about a year in order to give me the answers I was looking for. But even then, some challenges remained.

What follows are the most significant shortcomings of Cisco's AVVID/CallManager approach to enterprise VoIP, and recommends workarounds. This way, maybe, you can save some time by knowing which questions to ask your Cisco VAR, how to ask them, and how best to be prepared for their responses. It's aimed at people with a data networking background who are probably very comfortable with Cisco's datacom products, but who might not have a full understanding of telephony technology.