Think SIP trunking makes things simpler? Think again

It's easy to lose control if you don't forward plan and check around the edges

The general aim with SIP trunking tends to be around cost efficiency - bundling data, voice and video together down one pipe keeps phone and data bills down. That's standard enough. What's debatable is how much easier it'll make your life, and how much money you could actually lose if it doesn't work as smoothly as the marketing says it will.

New Computing research shows that 42 per cent of UK enterprise is now using SIP trunking specifically, while 52 per cent is using PBX - which probably includes SIP trunking too. This means most people are doing it, but does smashing everything together down one pipe automatically make things more straightforward?

As with anything, paying more will probably get you faster and more reliable quality, but at the end of the day, the configuration of your overall infrastructure is going to govern how well anything works when all your data is poured into the same spout.

And indeed, with such a high percentage of our respondents for the report Why SIP trunking is not enough using SIP trunking, it may be little surprise to find 32 per cent reporting quality of service problems, 28 per cent reporting call drops, 26 per cent unable to connect at all and 24 per cent reporting "jitter" on the line, it's not too difficult to draw a connection.

With SIP trunking, routing everything alongside everything else is going to quickly take up bandwidth, and that's likely to be your prime candidate for failure. Chucking more bandwidth at the problem also isn't always going to work as seamlessly as you may think:

"The pricing model employed by most SIP trunking providers can work against this kind of action, resulting in unexpectedly higher telecoms bills, especially on networks characterised by traffic peaks and troughs," advises the whitepaper.

Not to mention the fact that QoS problems can often lie outside your own network anyway - at the provider's end - meaning you may be paying for extra bandwidth that won't fix anything.

With 41 per cent of those responded saying either "No" or "Don't know" when asked if they've ever carried out service validation on their connections, and 24 further per cent saying they have, but not recently, it's all a potential recipe for disaster.

QoS and traffic prioritisation algorithms can help to auto-fix many problems, but still leave admins with less idea of what's going on than is ideal, while diagnostic ability pushed to you by providers can often not offer enough detail.

In fact, 47 per cent of those we asked feel they "could do with more detail" as to the visibility, understanding and control over their networks, with a further 14 per cent replying "Generally no" when asking if they have enough insight here.

To sum it up - there's no magic bullet approach to SIP trunking, and it's only ever going to be as good as the network it runs on. How good that network is depends on a variety of factors, many of which - inevitably - may emerge as being well beyond your control.

Plan carefully if adopting, and look into vast array of third party quality assurance and network monitoring software available, as the number of hidden pitfalls is only going to grow as we find more aspects of unified communications to shove down the same tube.

The Computing research paper "Why SIP trunking is not enough" is produced in association with NetScout, and will be published in full very soon. In the meantime, keep up to date with the latest news in unified communications and networking by visiting our UC&C Spotlight channel.