If there’s one thing that the comms industry likes to make a noise about, it’s resilience. The provider decks are as polished as can be and the case studies feature confident operations managers who make it all sound so straightforward. But when a capacitor goes on a core switch at twenty past three on a Tuesday afternoon, suddenly it’s a very different conversation.
As soon as your customers can’t get a call through, you try to move to Teams to accommodate but it’s also a bit wobbly because it was the same infrastructure carrying that traffic. At this point, someone in sales is putting together an apology email while another in engineering is drafting the incident report. By then, it’s usually too late and a frustrated customer has already posted about it online while others are checking when their contracts end.
Hardware failure is just one of those topics that likes to be kept theoretical. So what happens if it does become a reality even though you signed a contract on the back of a 99.999% uptime promise?
The Technical Failures That Don’t Make The Case Studies
When you hear “VoIP hardware”, you often think of just the handsets. And it is, but it’s also the routers, PoE infrastructure, the SBCs, the media gateways, the switches, the on-premise PBXs and NTUs at the network edge. Every single one of them has their own consequences of failure. For example, a dying SBC doesn’t look the same as a flaky PoE switch and knowing how to differentiate between issues will greatly affect your response time.
If anything has been learned the slow way, it’s that total failure is usually the easier situation to deal with. Dead kit announces itself, the monitoring picks it up and the standby does its job. But when you’ve got the half-broken kit, it can go unnoticed or leave you wondering if there’s a real problem or not.
The View From The Other End Of The Line
When something goes wrong on the carrier or hosted end, what the customer experiences doesn’t really match what you’re seeing on your side. When calls don’t go through or the audio quality is lost, it usually makes the caller assume that you’re being rude rather than facing a technical issue.
It gets even messier with on-prem because there are so many different parties involved. The kit belongs to the customer but the support contract belongs to the reseller and the firmware belongs to the vendor. When an incident arises, the first step is usually to figure out who’s responsible for it – that in itself is a major time-waster.
It’s often a long, mildly awkward stretch before any real work towards a resolution begins.
What It Costs Beyond The Uptime Sheet
A hardware failure rarely just affects uptime, it also makes a pretty decent dent in renewals. Even if a significant amount of time has passed, a customer tends to remember a bad outage.
Most of the frustration sits between what the contract says happened and what the customer feels has happened. Even if a provider were to hit every number in the Service Level Agreement (SLA), they could still lose the account because the customer remembers that frustration – very well.
Where Has The Risk Moved Now?
The move to hosted has played a major role in shifting the hardware risk onto providers and for end users, it’s broadly a good thing. On their side, there’s fewer things to worry about, less to maintain and less to go wrong on their own premises.
So the risk hasn’t really gone elsewhere, its just become more concentrated. And when something goes wrong at the platform level, it often doesn’t affect one customer – it affects many of them at once and that’s a different conversation entirely.
