Your sales team hears static on half their outbound calls. Customers keep saying “you’re breaking up.” Conference calls turn into a game of guess what they said. So IT replaces the handsets. Then someone blames the internet and upgrades the plan. Then the PBX cops it.

And the calls still sound terrible, because the actual problem is sitting one layer further up: the SIP trunk provider carrying your voice traffic, the one company in the chain nobody thought to question.

Bad call quality isn’t random

It feels random, which is why it survives so long. In practice it follows patterns, and they trace back to how cheap voice providers build their networks. They pile too many customers onto shared links. They bounce your calls through third-party carriers to shave termination costs. They don’t prioritise voice traffic, so when the link gets busy, your phone call loses the fight against somebody’s video stream.

The technical symptoms have names. Jitter: packets arriving out of time, which you hear as robotic voices and echo. Packet loss: gaps where words used to be. A little of either makes conversation tiring. A bit more makes it impossible. The congestion behind them peaks at exactly the times that matter, Monday mornings and end of month, when everyone is on the phone at once.

Why the cheap trunk is cheap

That bargain per-channel rate is achievable because the provider is betting not all their customers will use the capacity at the same time. When the bet fails, your call quality is what pays out. Add codec conversion to squeeze bandwidth, plus long routing paths through other people’s networks, and you get a service that demos beautifully at 2pm on a Tuesday and falls apart under real load.

The part that actually breaks people: the carrier maze

Here’s the scenario we get called into. The IT team has done its homework. They can see the packet loss, they know it’s upstream, they’ve raised the ticket with evidence attached. What happens next is a masterclass in deflection: the ticket lands with a first-level team that asks them to reboot the router. Then it’s escalated to a different team that asks for the same logs again. Then the voice team says it’s a network issue, the network team says it’s a voice issue, and week three arrives with the case reassigned to someone new who opens with “so what seems to be the problem?”

Unless you’re one of the carrier’s top 50 accounts, this is the service you’re getting. They’re not being malicious. You’re just small to them, and their support structure is built to exhaust small.

This is where we earn our keep. Getting carriers to actually fix things is a skill: knowing which team really owns the fault, what evidence forces an escalation instead of a loop, which SLA clauses have teeth, and who to call when the process stalls. We manage these fights every week across dozens of clients, so we know the shortcuts through each carrier’s maze, and carriers respond differently to a party that manages a book of their business accounts. Your IT team gets their week back, and the fault gets fixed instead of reworded.

And if the trunk itself is the problem

Sometimes escalation confirms what you suspected: the provider’s network is simply not good enough, and no amount of ticket wrangling will make it so. Then the fix is moving your voice to a properly engineered network, with quality-of-service prioritisation, direct carrier interconnects, and an uptime SLA written into the contract. We handle that transition too: the porting, the cutover, and the contract terms that stop this happening again.

If your team is stuck in a carrier support loop right now, or you’ve stopped raising the tickets because nothing ever comes back, get in touch. Fighting carriers is a service we offer precisely because so few businesses have the leverage to win those fights alone.