When you’re evaluating a push-to-talk platform, the hardware gets most of the attention. The devices, the button feel, the battery life, the ruggedness rating. That’s understandable — it’s the part your people interact with every day.

But the part that determines whether your comms are actually reliable isn’t the device. It’s the server infrastructure behind it. And for most PTT providers, that infrastructure is something they rent — from AWS, from Azure, from Google Cloud — rather than something they own.

This article explains what that distinction actually means in practice: for your uptime, for your data, and for organisations where communications going down isn’t just inconvenient.

What “Renting Cloud Infrastructure” Actually Means

When a software company says their platform runs on “the cloud”, what they usually mean is that they pay Amazon, Microsoft, or Google a monthly fee to use computing resources on shared data centres. The underlying hardware belongs to the cloud provider. The software company has no physical access to it, no ability to modify the hardware configuration, and no control over what happens if that third party has an incident.

For a lot of applications — a project management tool, a CRM, an invoicing platform — this is perfectly sensible. The major cloud platforms are reliable and the economics are good. There’s no compelling reason for a small SaaS company to buy and operate physical servers.

Communications infrastructure is different. When your PTT system goes down, it doesn’t mean someone can’t update a spreadsheet. It means your drivers can’t talk to base. Your security staff can’t reach their controller. Your dispatcher has no fleet visibility. The operational consequences of a comms outage are categorically different from most software failures.

“We own the hardware and the network. That means when something goes wrong, we own the problem too — and we can fix it directly, without waiting for a cloud provider’s support ticket to move.”

The SLA Numbers Nobody Explains Properly

Most cloud-hosted PTT providers publish uptime SLAs — a guaranteed minimum uptime percentage, usually expressed as a string of nines. These numbers sound impressive. Here’s what they actually permit in terms of downtime:

99.9%
Three nines — common SLA
8 hrs 46 min / year
99.99%
Four nines — “high availability”
52 min / year
100%
Press2Talk — 48+ months
0 min downtime

Downtime figures are per year. A 99.9% SLA is the most common published standard for SaaS platforms. “Five nines” (99.999%) — often cited as the gold standard — still permits over 5 minutes of downtime per year.

There’s another thing worth understanding about SLA percentages: they measure what’s allowed, not what actually happens. A provider with a 99.99% SLA can have a single 52-minute outage per year and technically still meet their contractual obligation. The SLA is a ceiling on liability, not a performance record.

Press2Talk doesn’t publish an SLA percentage because we track something more useful: actual uptime. As of April 2026, our network has recorded 100% uptime across 48+ consecutive months. That’s not a contractual floor — it’s a record.

Why this matters more for some industries

For a bakery coordinating delivery drivers, a 52-minute outage is an inconvenience. For a security company managing a lone worker who’s just pressed a duress alert, or a bus operator whose driver hasn’t checked in and the system is down, it’s something else entirely.

The value of genuine uptime scales with how critical your communications are. For regulated industries and operations with duty-of-care obligations, uptime isn’t a feature — it’s the whole point.

What Press2Talk Does Instead

Press2Talk operates its own physical server hardware, located in Australia. Not a virtual machine on someone else’s rack. Physical hardware that we specify, configure, maintain, and control.

Rented cloud infrastructure

  • Hardware owned by AWS / Azure / Google
  • Provider has no physical server access
  • Outages depend on third-party resolution
  • Data location may be overseas by default
  • Subject to third-party platform incidents
  • SLA covers liability, not actual performance
  • Configuration constrained by cloud provider

Press2Talk — owned infrastructure

  • Physical hardware owned by Press2Talk
  • Direct access — no intermediary
  • Issues resolved internally, without waiting
  • All data stays in Australia
  • No dependency on third-party platform health
  • 100% actual uptime — 48+ months on record
  • Configured specifically for PTT voice delivery

This isn’t a configuration choice we made to be different. It’s how we were built from the start. Our parent company has been in radiocommunications infrastructure since 2003 — the instinct to own and operate the critical hardware rather than outsource it is in the DNA of how this business works.

Want to understand our network in detail?

Our PTT network page covers the full technical picture — uptime record, encryption, data sovereignty.

See Our Network

Data Sovereignty — What It Actually Means

“Data sovereignty” gets used a lot in technology marketing. It’s worth being specific about what it means and why it matters.

When your voice communications pass through a server, they don’t just travel through it — they’re stored. PTT platforms typically record voice transmissions for audit, compliance, and playback purposes. Press2Talk retains voice recordings for 36 months. That’s a meaningful archive of your operational communications.

If those recordings are stored on a server in Virginia or Singapore or Frankfurt, a few things follow:

  • The data is subject to the laws of that jurisdiction, not Australian law
  • A foreign government or regulator could potentially compel access to it under their laws
  • Your organisation may have compliance obligations — under the Privacy Act, industry regulations, or client contracts — that require Australian data residency
  • In a security context, you may be required to demonstrate where your communications data is held

Australia’s Privacy Act requires organisations to take reasonable steps to protect personal information, including considering the destination of data when it’s transferred overseas (Australian Privacy Principle 8). For organisations in security, government supply chains, or healthcare, this isn’t a technicality — it’s an actual compliance requirement.

“Your data never leaves Australia. That’s not a policy we’ve written — it’s how the infrastructure is built. There is no overseas server for it to go to.”

Press2Talk’s servers are in Australia. Voice recordings are stored in Australia. AES-256 encryption is applied to data at rest and in transit. There is no third-party cloud provider through which Australian communications data passes.

Why This Matters More for Security Operations

Most of the above applies broadly to any fleet operation. But for security companies specifically — whether CIT, static guarding, mobile patrols, or event security — the infrastructure question becomes acute for a few additional reasons.

Security companies are increasingly asked by their clients to demonstrate data sovereignty as part of contract compliance. A major retailer or government agency awarding a guarding contract may require that communications infrastructure meets Australian data residency requirements. “We use a US cloud provider” is an answer that fails that requirement.

Voice recordings of security incidents have evidentiary value. Knowing that those recordings are stored locally, encrypted, and accessible under Australian legal frameworks rather than a foreign jurisdiction’s rules matters when something goes wrong and the recordings become relevant to an investigation or insurance claim.

And lone worker protection — duress alerts, man-down detection, GPS tracking — is only meaningful if the system is actually up when it’s needed. A security officer pressing a duress button during an incident is not well-served by a platform with a 99.9% uptime SLA that happens to be in its 8 hours of permitted downtime this year.

See how Press2Talk supports security operations specifically, and how the Web Dispatcher gives control rooms full visibility over field staff.

The Honest Version

Owning physical server infrastructure is more expensive and more operationally demanding than renting cloud capacity. It requires investment in hardware, in facilities, in maintenance, and in the expertise to run it. Most SaaS companies make the rational economic choice not to do it.

We make a different choice because our customers’ operations depend on the comms working. Not 99.9% of the time. All of the time.

If your communications are genuinely mission-critical — if people’s safety depends on the system being up, or if your compliance obligations require Australian data residency — then who owns the hardware your communications run on is not a background detail. It’s one of the most important questions you should be asking any PTT provider.

Our answer is: we do. And we can prove it.

Read more about our PTT network infrastructure or why organisations choose Press2Talk over the alternatives.