By Simon Woodhead
If you clicked on this post expecting an announcement that we’d thrown in the towel on our stance against dirty origin surcharges and are jumping on board with the latest sleazy money grab, don’t worry. In fact, we’re doubling down with a new feature, unique to Simwood as always.
For most customers, surcharges don’t exist – we don’t charge them and where they crop up on the odd call we just absorb them, relying on our advanced fraud and arbitrage protections to ensure we’re not a victim. That feels the right thing to do and continue doing.
For others, like international operators shipping traffic from a specific geography, we remove the complexity and the Russian Roulette from it. We can’t make the surcharges go away entirely, but we can be specific about what it is and provide a risk-free fixed price based on the profile of traffic. We also don’t aim to profit on any surcharge because that just feels dirty, so any premium is cost-oriented. This saves them accommodating OBR ladders or being fleeced by other international operators who want to charge 30c a minute when the actual surcharge at the interconnect level is 2c. Again, our fraud and arbitrage protections let us take the rough with the smooth in confidence.
However, there’s an edge case that we’ve handled with features announced today. What about regional operators who may have incumbency in multiple markets and want to send us traffic from all? Sure, we can blend profile and price, but that gets riskier all round. For every winner in their group, there’ll be a loser who is paying more than they otherwise would through us. Direct accounts for each don’t work either where, say, they’re in 10 markets but switch international voice from 2.
Our solution, which may also have other applications, is trunk-based a-number filtering, a-number-based routing and trunk-based pricing. Let’s run through these:
Trunk-based A-number filtering
We’ve already mentioned in our talk of KYC how certain customers might be given a trunk capable of originating +44 traffic, and another not. This is very similar. Take our example regional carrier; they could have a trunk for each operating subsidiary, with billing segregated by trunk, and each trunk restricted to certain a-number patterns. So the Nepal trunk will only allow traffic with a +977 a-number for example.
It follows that traffic originating from one country may have a different price to traffic originating from another. To do this in a risk-free way for our customers (avoiding a dirty OBR ladder), we simply have a tariff applied for a specific trunk now. Thus the Nepal trunk can have a completely different set of custom rates applied to the Ireland trunk, but those rates are specific and do not vary by a-number within the trunk. The a-number filtering ensures that France traffic is rejected for example, at least from these country specific trunks, eliminating the risk of surcharging.
It might be that our regional carrier can take control of trunk selection and either use specific authentication for each, or can set an X-Header on their central switches to override the trunk appropriately. It could be that they have no way of doing this but still want to take advantage of the above. Here, we can accept traffic on the default trunk and categorise it according to the a-number. In our example case, we’d segregate Ireland and Nepal, each into their own trunk billed at the appropriate rate.
What about the rest? Well we can optionally configure it to be allowed on the default trunk at the standard account rates, or rejected because it doesn’t match.
Combined, these three features give us unprecedented control to achieve commercially what nobody else in the market can in removing risk and passing through our economic advantages. Admittedly, if the colouring-in department weren’t on holiday, they would have had way sexier names, but I hope that doesn’t detract from understanding their power.
If these are relevant to you, please talk to us! We’ll look forward to speaking to many of you at Capacity Asia in Singapore in a few weeks.