By Simon Woodhead
There’s an odd phenomenon amongst international carriers that I’ve blogged about before but which seems relevant to mention again.
Big telcos have reputations built on their long established TDM networks, reputations for quality where both the quality and the economics benefitted from them having built out their physical interconnects around the globe. Today, customers still maintain that perception but almost invariably are connecting to them over IP, often over the public Internet. At the other end the big telco has purchased a number of magic boxes that enable this new VoIP thing, and they front their old network.
The problems come when their procurement or ‘business performance’ team discover that this magic box can also handle buying routes as well as selling. All of a sudden, that long-established brand perception of quality is represented by a magic box that is selling routes, purchased behind the scenes by ‘voice traders’ intent on minimising cost through a blended average quality. Who’s call is ever blended though? If you’re the poor sap going over the bad but cheap route that makes the numbers work, it’s your tough luck!
The result for them is a magic box that is super-profitable, and this well-known brand being diminished to a magic box buying goodness knows what routes from goodness knows where, plus the cost of legacy and pension fund deficit! I fail to see how that can be a long-term business model once buyers realise what they’re buying and that their brand association is entirely misplaced – any muppet anywhere can buy a magic box and consume cheap routes.
At Simwood, we’ve lost count of the times we’ve opened support tickets with the big operators we buy from that we connect with over TDM, only to receive back from them a SIP trace highlighting their misrepresented supply chain (they sold us their TDM footprint) and incompetence of magic-box operation in a single email. On other occasions we’ve reported very obviously analogue audio issues, only to receive a very delayed reply reporting 100% packet-loss to an IP address that doesn’t exist – but the problem is ‘definitely fixed’ they claim because they’ve deleted the IP address! More often than not though, there is no reply and it is often utterly pointless raising an issue.
Our approach is thus to conclude that there is no mythical tier-1 operator that can be relied on for any length of time to terminate globally across their own established footprint. Those that are good today, will likely do something daft tomorrow. We therefore maintain many relationships, route traffic based on quality metrics and remove them from route for specific countries, or in extreme cases entirely, when, and it is a ‘when’, they screw up. The result means you only need one Simwood and we can offer a consistently good, quality product.
After relatively consistent supply in 2017, we noticed an uptick in international issues in the fourth quarter across both the European tier-1s that we use for international traffic. Ironically these were more acute for European destinations! In one case they have been removed entirely and will remain so until we get a response that passes our BS filters, suggesting someone there has read the magic-box manual and understands how this world works.
We’re pleased though to be making a number of necessary changes to maintain our flexibility here as, to be frank, if all we ever do is remove supply for problem destinations we’ll eventually run out.
So, earlier in Q4 we re-introduced a previously very good European national incumbent that had gone bad in 2016. They seem to have learned the error of their ways and for now are giving us very few issues. Beyond that though we’re making further changes to add operators who we think still deserve the label tier-1, as they still have fibre in the ground and local interconnects that they seem to understand. Specifically:
- We’re adding in a national US ILEC to compliment our local CLEC connections. 75% of the traffic they transit completes on-net and they’re the terminating network for 50% of traffic, further enhancing our value-proposition in the US.
- We’re adding in a third European operator who actually has fibre in the ground across the continent and distinct regional business units. We’re interconnecting locally in each country, thus making our termination to European countries only one-step removed from us building our network there directly.
In both cases we’re in the late stages of interop testing and they will be introduced to routing in the very early new year. Hopefully, you won’t notice this and it’ll be like the many other changes we make every day to ensure we’re adding value for you, rather than trading on our history or blindly reselling muppetry!
