Beating FUD with MOS

David Byrd : Raven Call
David Byrd
David Byrd is the Founder and Chief Creative Officer for Raven Guru Marketing. Previously, he was the CMO and EVP of Sales for CloudRoute. Prior to CloudRoute, He was CMO at ANPI, CMO & EVP of Sales at Broadvox, VP of channels and Alliances for Telcordia and Director of eBusiness development with i2 Technologies.He has also held executive positions with Planet Hollywood Online, Hewlett-Packard, Tandem Computers, Sprint and Ericsson.
| Raven Guru Marketing

Beating FUD with MOS

My interest in the Mean Opinion Score (MOS) increased when a Verizon spokes person in 2008 stated that their VoIP product was not as good as their TDM or traditional voice product. The comment came as result of questioning by financial analysts regarding the future of their TDM network and timing of the transition to an all IP network. It is true that the current reliability of IP networks is just shy of TDM networks (99,995 vs. 99.999). I will not quibble over the 21-minute difference, it is real and each business must decide if it should influence their decision to use VoIP/SIP Trunking as their primary method of communications. What I did find puzzling was the admission of reduced voice quality and the fact that they charged the same for VoIP as traditional phone service. Clearly, Verizon doesn't want to have customers move off their traditional network in droves, hence the pricing strategy. However, the voice quality statement is meant to influence SMBs and enterprises into standing pat, while consumers wanting to save money and early adopters experiment with VoIP offerings. Pure and simple this is freezing a market with FUD (Fear, Uncertainty and Doubt) until you are ready to make your move.

When I discovered MOS, I knew I had a compelling argument. Here was a method of comparing voice quality by independent means where VoIP and toll calls were deemed equal. On a scale of 1-5 where 5 is perfect, VoIP with a G.711 codec and a toll call came out equal with 4.4 each. So how is the measurement done?

To establish a MOS, listeners (in the case of the study I am quoting 30 to 50) listen to speakers of both sexes read various sentences. Each listens rates each sentence as follows: (1) bad; (2) poor; (3) fair; (4) good; (5) excellent. The arithmetic mean of all the individual scores is the MOS. The International Telecommunication Union (ITU) endorses testing method is endorsed by. Here are some of the suggested sentences:

·         You will have to be very quiet.

·         There was nothing to be seen.

·         They worshipped wooden idols.

·         I want a minute with the inspector.

·         Did he need any money?

Obviously, the results of a MOS test are subjective. Therefore, software tools have been developed that automate MOS testing. These tools do a better job of evaluating network conditions and codecs more consistently than human listeners.


So, forget the FUD, get out there and let people know. VoIP/SIP Trunking is as good as your current phone service and with Broadvox and other ITSPs, you can save up to 70%.


See you on Monday.

Related Articles to 'Beating FUD with MOS'
Feedback for Beating FUD with MOS

1 Comment

The problem with MOS is that it is only a symptom of deeper-level network problems. Many folks get too entrenched in looking at and analyzing MOS scores and don't realize that the important task is to resolve the problem.

MOS's failing is that it has no ability to get to the root-cause of the problem. It will never be able to disclose a duplex mismatch on a switch trunk port, or a cable fault on a station cable.

You need total network visibility to find and resolve the problem.

Otherwise, you're just left with a MOS score of 2.5 and a frustrated user because you have no information that's helpful in getting the problem resolved, only confirmation that a problem exists.

Here's a best-practices way of finding & resolving the root-causes of VoIP issues:

Flash: Finding & Resolving VoIP issues

Leave a comment

Featured Events