The NBN did its job. It replaced the copper network, and for households it was a genuine upgrade. For businesses it’s often the wrong product, sold by default because it’s what’s available and what’s cheap.

We see the result in audits constantly: a company that has grown from eight staff to forty, still on the NBN plan they signed years ago, with degraded calls and crawling uploads treated as facts of life. Nobody reviews the connection until it becomes unbearable, and by then it’s usually been costing money for years.

The problem isn’t download speed

NBN business plans advertise downloads up to 1,000 Mbps, which sounds like plenty. Look at the upload figure instead. On most of those plans it’s 20 to 50 Mbps, shared with every application you run.

A modern business hammers its upload far harder than its download. Cloud backups go up. Files sync up. Your side of every video call and every VoIP conversation goes up. When that narrow upstream pipe fills, everything degrades at once, and it usually fills in the middle of the working day.

Here’s the part that surprises people: there is nothing you can do about it. NBN plans carry no guarantee on upload performance and no remediation when it underdelivers. Speeds are quoted as “typical”, contention with the neighbourhood is by design, and if the connection fails, repair is best-effort with no committed timeframe. There’s no SLA to enforce and no one to escalate to. That’s not a flaw in your plan; it’s the product.

What dedicated fibre actually buys

Dedicated business fibre (also sold as Enterprise Fibre or Ethernet over Fibre) is a different product, not a faster tier of the same one. The speeds are symmetric, so 500/500 means 500 Mbps in both directions. The bandwidth is yours alone, uncontended, around the clock. And the SLA is contractual: uptime guarantees around 99.95%, restoration targets measured in hours, and penalties when the carrier misses them. You’re buying recourse as much as speed.

To make the speed difference concrete: a 5GB file that takes fifteen minutes to upload on a 50 Mbps NBN upstream takes about ninety seconds on 500 Mbps fibre.

When NBN is genuinely fine

We’re not in the business of selling fibre to people who don’t need it. A small office, light cloud usage, email and browsing with the occasional video call: NBN handles that fine, and paying for dedicated fibre would be exactly the kind of over-provisioning we get paid to remove. NBN Enterprise Ethernet also exists as a middle option, with symmetric speeds over NBN infrastructure at a price point between the two. It’s one of several connectivity services we supply and manage where it’s the right fit.

The signals you’ve outgrown it

• Multiple sites on VPNs. Every site’s traffic funnels through the upload of the hub. Asymmetric plans put a choke point exactly where you can least afford one.
• Your business lives in the cloud. If your files, applications and backups sit in AWS, Azure or Microsoft 365, upload speed is effectively your office speed.
• Phones and video run over the connection. Ten concurrent HD calls want 30 to 50 Mbps of clean, jitter-free upstream. That’s most of an NBN plan’s upload gone before anyone opens a file.
• Real-time systems. POS terminals, medical imaging, anything where a lag spike costs money or safety. These need consistent low latency, which shared connections don’t promise.

What it costs, honestly

Dedicated 500/500 fibre generally runs somewhere between $550 and $750 a month depending on location and term, with gigabit services from around $900. Fibre now reaches tens of thousands of Australian business premises, and where it doesn’t, carriers will often build it in for the term of a contract.

Whether that’s worth it depends on your workload, which is a calculation we’d rather do properly than assert in a blog post. The fibre decision is also a favourite spot for overspending in the other direction: businesses paying for gigabit symmetric fibre to run email. We’ve unwound plenty of those too.

Get the decision made properly

Your internet connection is infrastructure, and infrastructure decisions deserve better than a guess. As part of a Telco Health Check we benchmark what you’re on against what your workload actually needs, in both directions. Sometimes the answer is an upgrade. Sometimes it’s a cheaper plan. Either way you’ll know, and it costs you nothing to find out.

Ask how much bandwidth your organisation needs and someone will produce the formula: count the users, multiply by some Mbps per head, add 20% for luck. It fits on a whiteboard, it feels rigorous, and across the enterprises we audit it is reliably wrong in one of two expensive directions. Either the site is undersized and everything crawls at peak, or it’s oversized and finance has been paying for headroom nobody will ever touch.

At enterprise scale, both errors multiply. Ten oversized sites is a six-figure annual habit. Ten undersized ones is a productivity problem that gets misdiagnosed as an IT problem for years.

Users don’t consume bandwidth. Workloads do.

A 20-person architectural practice and a 20-person call centre have almost nothing in common on the wire. The architects push 10GB model files to the cloud and sit in video reviews half the day. The call centre needs a modest but immaculate stream of low-latency voice where a single congested moment is an abandoned customer. Head-count formulas treat these as identical businesses.

Sizing that holds up starts from the application mix, not the org chart. What actually rides this connection: cloud backups, video, voice, file sync, remote desktop, security cameras? When does each run, how much does it burst, and how badly does it tolerate delay? Do that mapping properly and interesting things fall out. That backup job demanding 200 Mbps runs at 2am and shouldn’t be sized into the daytime peak at all. The daytime number that matters might be barely half of what the formula said.

The three questions that actually size a connection

How much contention can you absorb? If speeds sagged 30% at peak, would your teams notice and cope, or would you lose revenue? The honest answer decides between shared infrastructure and uncontended fibre, and it’s different for a head office than for a distribution centre. Paying for uncontended fibre at a site that would happily cope on less is one of the most common oversights we unwind.

What does an hour of outage cost? Uptime percentages sound interchangeable and are anything but. 99% allows three and a half days of outage a year. 99.9% allows nearly nine hours. 99.95% allows about four. Somewhere in your organisation is a site where four hours offline is a shrug, and another where it’s a board paper. They should not be on the same class of service, and under most enterprise agreements we review, they are.

Where will you be in two years? Standing still, enterprise bandwidth demand grows 20-30% a year. Add a cloud migration or a serious video culture and it’s 50% or more. Connections should be sized for 18 to 24 months out, on contracts with an upgrade path that doesn’t require re-engineering the site. Otherwise today’s correctly sized link becomes next year’s bottleneck, locked in for another two years.

The part the carriers won’t do for you

None of this analysis comes from your provider, for the obvious reason: the carrier’s incentive is to sell you the bigger number. The formula suits them fine. What an enterprise actually needs is the workload map, site by site, matched against what each site is currently paying, and the mismatches priced. In our experience the same review that finds the undersized head office also finds three branch sites on enterprise-grade services running email and a point-of-sale terminal.

That review is squarely what a Telco Health Check covers. We map what your sites actually use, benchmark it against what you’re contracted for in both directions, and hand you the corrections. Sometimes that’s an upgrade. More often than the carriers would like, it’s not.

If your business runs more than a handful of sites, your telco spend has a particular shape, and we’d bet money on it before opening a single invoice. Head office on decent fibre. The warehouse on NBN. A regional branch limping along on whatever was available the year it opened. Each site got its connection when it got it, from whoever could deliver, and nobody has reviewed the whole picture since.

Multi-site businesses are where we find the biggest telco waste, consistently. Not because anyone was careless, but because connectivity bought site-by-site over a decade produces a portfolio nobody designed and nobody owns.

The MPLS legacy tax

The biggest single offender is the legacy MPLS network. For twenty years MPLS was the respectable enterprise answer for connecting sites: private links, guaranteed bandwidth, and prices set in an era when there was no alternative. A 100 Mbps link between two capitals can run $1,500 to $2,500 a month. Across ten sites that’s a quarter of a million dollars a year, spent before anyone opens a browser.

Businesses keep paying it for two reasons. The contract auto-renews, and the network works, so nobody wants to touch it. Both are understandable. Neither changes the fact that equivalent reliability is now available for 60-70% less, because SD-WAN over modern fibre does what MPLS did at consumer-adjacent link prices. For that same ten-site business, the modern equivalent typically lands between $95,000 and $140,000 a year, with better cloud performance and new sites connected in days instead of the 8 to 12 weeks an MPLS addition takes.

If your MPLS contract predates 2020 and hasn’t been to market since, you are almost certainly paying the legacy tax.

The other leaks multi-site creates

• Site-grade mismatch. Enterprise-grade fibre at a depot with three staff, and contended NBN at the branch doing a third of company revenue. Allocations set by history, not by what each site earns or risks.
• Redundant redundancy. Backup links at sites that could tolerate a day offline, no backup at sites that couldn’t tolerate an hour.
• Orphaned links. Sites close and consolidate; their data services keep billing. The more sites, the more closures, the more orphans. On consolidated invoices spanning dozens of locations, a dead link for a closed site is invisible.
• Carrier sprawl. Five sites, four carriers, four bills, four support numbers, and no single rate negotiation across any of it. Every carrier prices you as a small customer because to each of them, you are.

Why nobody inside can see it

The information needed to fix this sits in three different places. Finance has the invoices but can’t tell an essential link from a dead one. IT knows what each site needs but doesn’t see the bills. And the contracts live wherever the person who signed them filed them, possibly before two restructures ago. Assembling the full picture (every site, every service, every contract, every dollar) is real work, which is why it stays unassembled, and why the waste compounds quietly year after year.

Carriers understand this perfectly. Multi-site accounts on legacy contracts are among their most profitable business, precisely because the customer can’t easily see the whole board.

Seeing the whole board

This assembly job is the core of what a Telco Health Check does for a multi-site business: every location, every service and every contract in one view, matched against what each site actually needs and what the market now charges. The outcome is a list of concrete corrections, from dead links to cancel this week, to the MPLS renegotiation worth six figures over its term. For multi-site operations the findings are rarely subtle. There are simply too many places for money to hide, and too many years since anyone looked.

Hybrid work settled in years ago, but the telco spend it created never really got settled at all. Home internet stipends approved during 2020 and never reviewed. Mobile broadband dongles issued to staff who’ve since changed roles, or left. Data plans upgraded for a project that finished. A video subscription here, a softphone licence there. Individually small, scattered across expense reports and a dozen cost centres, and in most businesses, owned by nobody.

That’s the first thing worth saying about remote workforce connectivity: it’s not just an IT setup question, it’s a spend category now, and it has all the habits of one that’s never been governed. When we audit mobile fleets, the remote-work layer is reliably where the strangest artefacts live.

The productivity side is real too

None of that means skimping. A remote worker on a struggling home connection loses real hours: choppy meetings, file uploads that stall, VPN sessions that drop. Home internet that’s fine for streaming is frequently not fine for eight hours of video calls, and the member of staff usually won’t complain until it’s cost them a client interaction.

The fixes are established: business-grade plans for key staff, mobile broadband as backup for the roles that can’t afford an outage, and tools that work the same from a kitchen table as from the office. One consultancy we work with equipped its senior remote staff with dedicated 4G backup routers after tallying the billable hours lost to home internet outages; the gear paid for itself inside two months.

On tools, the direction of travel is clear: calling, meetings and chat inside the one platform staff already use, so a remote worker’s “desk phone” is their laptop and mobile, with no separate systems to maintain. Security follows the same logic. Distributed teams need multi-factor authentication everywhere, managed devices, and access controls designed for people who never touch the office network.

Provision deliberately, review on a cycle

The failure mode isn’t usually the technology. It’s that remote provisioning happens ad hoc (a request, an approval, a recurring cost) and then nothing is ever revisited. The stipend outlives the circumstances. The backup SIM outlives the employee. Multiply by a few years of staff turnover and the remote-work layer of your telco spend becomes a little museum of past arrangements.

Treat it like any other fleet instead: a register of who has what and why, plans matched to actual usage, and a review cycle that catches the leavers and the role changes. That’s standard mobile fleet governance extended to where people actually work now.

If nobody in your business could currently produce the list of remote-work connectivity you’re paying for, that’s a solid sign it’s time someone looked. A Telco Health Check covers the remote and mobile layer alongside everything else: what exists, what it costs, what’s still earning its keep, and what should have been cancelled when its owner handed back their laptop.

Run a speed test at your office and the download number will look great. It’s the number the plan was sold on. Now watch what happens when the design team uploads a 4GB file, or the nightly cloud backup is still running at 9am, or a video call freezes the moment someone shares their screen. None of that is the download’s fault. It’s the other number, the one in small print: on a typical business NBN plan, upload is 25 to 50 Mbps against a download twenty times wider.

For a household, that asymmetry is sensible; homes mostly consume. A business mostly produces, and everything it produces goes up.

Where upstream bandwidth actually goes

• Cloud backups. A 50-person business generates hundreds of gigabytes of new and changed data weekly. At 50 Mbps up, a serious backup takes 9 to 22 hours and starts colliding with the working day. At 500 Mbps symmetric it’s done in under two.
• Video calls. Each 1080p call sends 3 to 5 Mbps upstream, more with screen sharing. Ten concurrent calls want 50 to 90 Mbps of clean upload. On a 50 Mbps plan that’s the whole pipe, before a single file moves.
• File sync. OneDrive, SharePoint and Dropbox can’t distribute a file to the team until the original finishes uploading. One slow upstream throttles everyone downstream of it, all day.
• Voice. Every VoIP call rides the upload too, and voice is the least forgiving traffic you have. When backups and video have already saturated the link, call quality is the first casualty.
• Remote access. Staff connecting back to office systems consume upload at the office end. A few remote workers on demanding applications will feel a starved upstream before anyone on site does.

Add those up for your organisation and compare the total to the upload figure on your current plan. For most businesses past a certain size, the arithmetic simply doesn’t close, and the daily symptoms (the stalls, the freezes, the overnight jobs that aren’t finished by morning) are that arithmetic expressing itself.

What symmetric actually means

Symmetric services (500/500, 1000/1000) give you the same capacity in both directions, and on proper business fibre that capacity is uncontended and covered by an SLA. The catch is that symmetric is a different product tier, not a toggle on your current plan, and it costs accordingly. Which makes this a sizing decision: real workload, both directions, priced against the productivity being lost.

It cuts the other way too. We’ve audited businesses paying for gigabit symmetric fibre whose upstream never exceeds a trickle. Upload is where under-provisioning hurts and over-provisioning hides, and both mistakes are common because almost nobody measures the upstream side before buying.

Measure before you buy

A Telco Health Check looks at connectivity in both directions: what your sites actually push upstream, whether the current plans can carry it, and whether you’re paying for headroom you’ll never use. The download number takes care of itself. It’s the other one that decides how your business runs.