Lately we’ve been talking a lot about quality in relation to voice services. How to measure it, what you can measure and even why you should be doing so. I won’t retread the same ground here. Suffice to say that, in an age of increasingly commoditized VoIP services, CX becomes a major differentiating factor, so ensuring consistently high quality on your phone lines is of paramount importance.
But what does that really mean, ‘ensuring consistently high quality’? Or to put it another way: If you test the quality of your lines and the result shows them to be wanting, what exactly can you do about it? To answer that question, you’ll first need to consider precisely where in the call journey that performance is dropping. After all, you can’t troubleshoot a problem until you’ve isolated it.
There are three main areas where issues can occur that will harm call quality:
- In your own network
- In your voice provider’s network or downstream
- The interconnection between 1 and 2
We’ll take a look at each of these in turn. But first, let’s think for a moment about how you can accurately peg a quality issue to one of these three stages in the call journey.
Isolating Quality Issues in Real Time
How do you tell where poor performance is entering your call flow if it isn’t the result of something starkly apparent, such as a provider network outage?
Up until now, this has been very time consuming and very expensive to do in real time, relying upon a combination of carrier metrics (if even available) and third-party solutions that usually entail some sort of probe to be implemented within your own network.
So it’s not surprising that all but the most devoted of call quality evangelists either fall back to relying on sporadic automated number testing or don’t bother at all.
A better way has arrived. With Voxbone Insights, our new call analytics platform, you get a complete real-time view of performance, from caller to callee – across metrics including mean opinion score (MOS), jitter, packet loss and round-trip time (RTT).
Having these metrics in real-time on every call is already great. But where Insights really shines is in how it lets you drill into these numbers and split them into two different call journey phases:
- Phase 1: Covering the Voxbone network and downstream providers or PSTN infrastructure
- Phase 2: Covering your network and the path between it and the Voxbone network
The beauty of this split is that, in essence, the first phase is our responsibility to maintain and improve, providing a quantifiable proof point by which our customers can judge our quality of service. And the second is for you to control, with our support and advice of course.
Voxbone Insights beta is live now. It is due for launch in September. Learn more here.
1. Troubleshooting Quality Issues Affecting Your Own Network
One of the major benefits of switching to cloud-based SIP trunking services is that you no longer need to maintain separate PSTN connections from your business.
By working with an operator capable of providing full PSTN replacement from the cloud, your voice operations can be run entirely over a data network. And your IT team can put in place traffic management policies to prioritise real-time media including voice and video so that network congestion doesn’t harm the quality of your phone calls and conferences.
Ultimately, at this stage of the call journey, it’s in the hands of your IT team or network administrator to ensure that everything is set up for optimal quality.
2. Tackling Poor Quality From Your Provider & Beyond
Most of the call journey will be the responsibility of your voice provider to guarantee. If you’re working with reputable operators that have their own in-country infrastructure and number ranges – or source them directly from a Tier-1 carrier where regulations make it difficult for external providers to operate directly – then this becomes much simpler.
If there are issues with a country’s national infrastructure, it’s likely to affect all operators equally and there is little a provider can do beyond pushing for a swift resolution and keeping you informed.
But if your provider aggregates voice coverage from unreliable carriers or is regularly affected by network downtime and loss of coverage, that’s the point at which you should consider looking for a different provider.
3. Squashing Quality Issues That Pop Up In Between
Assuming your IT team knows what it’s doing and you’re working with a decent voice provider, the most likely cause of quality issues is actually the interconnection between your network and your provider’s.
The number one culprit here is the internet. If you don’t have a dedicated private connection – whether through an SD-WAN provider like Megaport or a direct physical cross connect within a datacenter – your calls will have to traverse the public internet for this leg of the journey.
While the internet offers the broadest possible connectivity, enabling you to send and receive calls from any location, its public nature can have serious negative repercussions for the quality of real-time media including voice and video.
Because routing can be so unpredictable and your traffic is in contention with literally everyone else’s, not to mention the impact that traffic management policies can have on delivery. High latencies, packet loss and jitter are all much more likely on the open web.
Another major issue with the public internet is security. Encryption is one solution, but bear in mind that this will actually further deteriorate the quality of your calls by adding additional latency in the form of encryption and decryption times, not to mention the risk of higher levels of jitter.
Private Interconnections Close to the Source
What can you do to get your call traffic off the public internet and start improving quality? By establishing a dedicated private link between your network and your supplier’s, you can cut out a lot of the jitter and packet loss you’ll encounter online, as well as bringing latencies down and improving security. Albeit at a greater financial cost.
But one thing to always keep in mind is that latency is based on the laws of physics. The further a distance that a signal has to travel, the longer it will take. So even with a private interconnection, if you’re sending data halfway around the world, there’s still going to be more latency.
That’s why we have architected our network to provide low-latency coverage across America, Europe and APAC by positioning our SuperPoPs in geostrategic locations throughout these continents.
And if that’s not enough, all of these core locations are accessible via any of the 500+ locations on Megaport’s private global network, which provides scalable bandwidth for any sort of single or multi-cloud setup.
The interconnection choice that’s right for you will depend largely on your network makeup. But all of the connection types available as part of Voxbone Connect can be easily mixed and matched to suit your connectivity needs in each discrete location where you operate.
We’ll be talking more in the coming months about how the right connectivity can improve quality AND cut costs by including built-in resiliency without the need for backup channels.