The ACCC Inquiry: What It Is and Isn't
In August 2026 the ACCC commenced an inquiry into mobile services. Its stated purpose is to examine regional mobile coverage across Australia and to consider whether regulating wholesale access to some mobile services would improve outcomes for consumers.
The specific question in front of it is whether to declare one or more wholesale mobile services. Two candidates are named.
Domestic roaming
A declaration would oblige networks to carry each other's customers where the customer's own network has no coverage. Your phone would use whichever network is present, rather than showing no service because your provider is absent.
A radio access network service
A different mechanism — access to the tower and radio equipment themselves, allowing a provider to run its own service over another operator's infrastructure in areas where it has none.
The inquiry will run for approximately twelve months. A discussion paper is to be released shortly, and the ACCC has said it will hold forums in regional and remote Australia to hear directly from communities. Two contextual factors were explicitly weighed in the decision to proceed: the proposed introduction of the Universal Outdoor Mobile Obligation in 2027, and the ACMA's upcoming 2 GHz spectrum allocation.
Read the timeline before you plan around it
Twelve months of inquiry produces a decision about whether to declare, not coverage. If a declaration follows, terms and pricing have to be settled, then systems built, then services launched. The realistic distance between today and a regional business seeing a practical difference is measured in years, not months. Anyone selling you a connectivity decision on the basis that roaming is about to be mandated is selling you a decision you should make on today's facts instead.
The positions are exactly where you would expect. Telstra opposes mandated roaming, arguing it weakens the incentive to invest and therefore produces worse long-term outcomes for regional Australia — the network with the largest regional footprint has the most to lose from being obliged to share it. TPG, the smallest of the three mobile network operators and the most likely beneficiary of regulated access, supports the inquiry as an opportunity to improve regional competition. Optus has welcomed the opportunity to contribute. None of this is surprising, and none of it tells you what the ACCC will decide.
Why This Was Refused Before
This is not the first time. The ACCC ran a domestic mobile roaming declaration inquiry in 2017 and decided not to declare. Understanding that decision is the fastest way to judge how likely the answer is to differ this time.
The 2017 reasoning turned on investment. Compelling access at regulated terms reduces the return on building regional infrastructure, and if the return falls, less gets built. The ACCC was not persuaded that mandated roaming would deliver enough benefit to outweigh that effect.
Several things have changed since, and it is worth being clear about which are genuinely relevant to the question the ACCC has to answer.
| What has changed since 2017 | Does it bear on the declaration test? |
|---|---|
| The 3G shutdown removed a coverage layer that had been carrying rural traffic for two decades | ✓ Yes. It materially changed the real-world coverage picture in a way the 2017 analysis could not have anticipated |
| Major outages and their consequences, including emergency call failures | ~ Partly. It sharpens the resilience argument — a second network reachable during a failure — though roaming is designed for absence of coverage, not for outages |
| Satellite direct-to-mobile now exists commercially | ~ Cuts both ways. It is a competing answer to the same problem, and a regulator may reasonably ask whether the market is solving it already |
| A proposed Universal Outdoor Mobile Obligation from 2027 | ✓ Directly. If an obligation is about to deliver outdoor coverage by another route, the case for declaring roaming is weaker, and the ACCC has said it took this into account |
| Substantially higher spectrum renewal costs for carriers | ✗ Not directly, though it strengthens the industry argument that investment incentives are already under pressure |
The honest handicap
The considerations pull in both directions, and the 2027 obligation is genuinely awkward for the roaming case: it is difficult to argue that access must be mandated to fix outdoor coverage while a separate obligation to deliver outdoor coverage is being introduced. Plan on the basis that nothing changes, and treat a declaration as upside. That is not pessimism; it is the only planning posture that leaves you better off under either outcome.
Satellite Texting: The Honest Limits
This is the development that has generated the most excitement and the most confusion, so precision matters.
Telstra became the first Australian carrier to launch direct-to-mobile satellite services, working with Starlink. A compatible handset, with no terrestrial coverage at all, can send and receive text messages. In July 2026 this extended beyond texting to a curated set of satellite-optimised applications — messaging, navigation, weather. Coverage spans mainland Australia and Tasmania, excluding the Australian Radio Quiet Zone in Western Australia. Optus has indicated a comparable service during 2026, subject to regulatory approvals.
That is a genuine milestone. Now the limits, stated plainly, because they determine whether this is relevant to your phone system at all.
| Capability | Status |
|---|---|
| Text messages | ✓ Yes, on compatible handsets |
| A curated set of satellite-optimised apps | ✓ Yes, since July 2026 |
| Voice calls | ✗ No |
| General mobile data | ✗ No |
| MMS and ordinary internet messaging apps | ✗ No |
| Indoors, in a vehicle, under trees | ✗ Needs to be outdoors with a clear view of the sky |
| Any handset | ✗ Compatible devices only |
What this means for a business phone system
Nothing, yet — and that is fine. Direct-to-mobile satellite is a safety and coordination tool: a worker in a paddock can confirm they are alright, a driver can report a delay, a crew can be redirected. That is valuable and it is not what your business number does. Treat it as a category of its own, sitting alongside the phone system rather than inside it, and do not let it substitute for the reachability work below. Note also that it does not carry a voice call to Triple Zero — a point worth making explicitly to any field staff who assume otherwise. Our Triple Zero and cloud phones guide covers the emergency calling rules properly.
The trajectory is real. Direct-to-mobile voice is being worked on internationally and will arrive. When it does it will change the calculus for remote operations considerably. It has not arrived, and a business plan that assumes it this financial year is a plan with a hole in it.
The 2027 Obligation and the Spectrum Question
The Universal Outdoor Mobile Obligation is the third moving piece, proposed for introduction in 2027. In concept it obliges the delivery of outdoor mobile coverage across a defined footprint, with satellite technology anticipated as part of how that gets achieved in the hardest areas.
Two observations that matter more than the detail, which is still being settled.
“Outdoor” is the operative word. An obligation framed around outdoor coverage does not promise a usable signal inside a steel shed, a cool room, a workshop or a homestead with a metal roof. For a very large number of regional businesses, the coverage problem is indoors at their own premises, and that problem is not addressed by any of the three developments on this page. It is addressed by putting your voice on an internet connection you control — which is the entire argument of the second half of this article.
Spectrum costs shape what gets built. The ACMA's forthcoming 2 GHz allocation was cited by the ACCC as relevant context, and carriers are facing substantial spectrum renewal costs. Spectrum economics determine the pace of regional network investment more directly than most policy debates do, and they are the quiet variable behind every argument in this space.
The Four Ways Coverage Actually Fails You
Regional connectivity problems get discussed as one thing — “no coverage” — and they are four different things with four different fixes. Sorting your own situation into the right box is most of the work.
| Failure | What it looks like | What actually fixes it |
|---|---|---|
| No coverage at the premises | The site has poor or no mobile signal, but has fixed broadband or satellite internet | Voice over the internet connection. Desk handsets and apps work fine; the mobile network is irrelevant to them |
| Indoor blackspot | Signal outside, nothing in the shed or workshop where people actually work | Wi-Fi. A phone system app calls over Wi-Fi like any other internet application. This is the most common and most fixable case |
| Coverage in the field | Staff genuinely outside any network, in vehicles or on properties | Nothing carries a voice call here today. Satellite messaging, scheduled check-ins, and a phone system that captures what they miss |
| The link itself fails | The connection at the site drops — weather, power, a cut fibre, an outage | Layered failover, decided in advance and tested. This is the one businesses assume is handled and usually is not |
The distinction worth internalising
Coverage is about signal. Reachability is about whether a customer who rings your number gets a human. They are not the same, and only the second one is your business's problem. A site with no mobile coverage at all can be perfectly reachable. A site with four bars can miss every call after 4pm. The regulatory debate is entirely about coverage. Everything you control is about reachability.
What to Build Now
Five design decisions. None of them require anything that does not exist today, and none depend on the outcome of an inquiry.
| # | Decision | Why it matters regionally |
|---|---|---|
| 1 | The number lives in the cloud, not at the site | A number tied to a physical line at your premises fails when the premises fail. A number that terminates in the network can be redirected in seconds from a phone in a car park. This single change does more for regional resilience than any other item here |
| 2 | Failover is defined in advance, not improvised | Write the order down: primary link, secondary link, staff mobile app, divert to another site or number. Configure it, then leave it. The alternative is a person making decisions during an outage, which is when people are least available |
| 3 | The mobile app is a full extension, not a compromise | Field and remote staff should place and receive calls on the business number from wherever they have any internet — the depot Wi-Fi, a home connection, a hotel. Personal mobiles making personal-number calls is a records problem and a continuity problem |
| 4 | Wi-Fi calling is treated as core infrastructure | The indoor blackspot is the most common regional complaint and the cheapest to solve. Access points in the shed, the workshop and the cool room convert dead zones into full-quality voice |
| 5 | Nothing is lost when nobody answers | Out of range is a fact of regional life. Missed calls logged and visible, voicemail transcribed to text, an SMS auto-reply, and a callback list mean an unanswerable call still becomes a returned one |
Point five deserves more weight than it usually gets. Regional businesses lose more revenue to calls that were never returned than to calls that could not be connected — the same pattern documented in what missed calls cost tradies. A caller who reaches voicemail and gets a call back within the hour is a retained customer. A caller who reaches nothing is a competitor's customer.
Choosing Your Internet Paths
Voice is a light load and a fussy one. It needs very little bandwidth and cares a great deal about consistency, which inverts the usual assumptions about what makes a connection good.
~100 kbps
Per concurrent call, each way
<150 ms
One-way latency target
<30 ms
Jitter — the number that actually breaks calls
<1%
Packet loss
Ten simultaneous calls is about one megabit each way. Almost any working connection can carry that. What ruins a call is not insufficient speed but variable delay, which is why a fast link with erratic behaviour sounds worse than a modest, steady one.
| Path | Suitability for voice |
|---|---|
| Fixed-line broadband, fibre or fixed wireless | ✓ First choice where available. Predictable and easily sufficient |
| 4G/5G fixed wireless | ✓ Good primary or secondary. Check the site's own performance at 8am and 5pm, not the coverage map |
| Low earth orbit satellite broadband | ✓ Workable for voice. Latency is in the range terrestrial links occupy; brief interruptions during handover are the practical limitation |
| Geostationary satellite broadband | ✗ Poor for voice. The round trip to a satellite ~36,000 km up imposes a delay no configuration can remove, producing the talk-over effect familiar from old satellite phone calls |
| Direct-to-mobile satellite | ✗ Not applicable. Text only, as covered above |
The orbit distinction is the single most useful technical fact for a remote site, and it is routinely missed. “Satellite internet” describes two products with completely different voice behaviour. If a site is on satellite broadband, establish which kind before deciding whether voice can run over it. Our VoIP and SIP troubleshooting guide covers diagnosing a link that is fast on paper and poor in practice.
The Field Test That Tells the Truth
Coverage maps describe modelled outdoor signal. They do not describe your shed. Half a day of testing produces a more accurate picture of your business than any map, and gives you something to hand a provider.
| Test | What to do | What it reveals |
|---|---|---|
| Where people actually stand | Place a call from every location a person works: office, shed, workshop, cool room, yard, the far paddock gate | The real map, which never matches the published one |
| Every carrier, not just yours | Repeat with a handset on each of the three networks. Borrow them for a day | Whether your problem is coverage or your provider. Often it is the second |
| Wi-Fi call quality | From the same locations, call over site Wi-Fi using the phone system app | How many blackspots an access point would eliminate. Usually most of them |
| Pull the plug | Physically disconnect the primary internet during business hours and watch what happens to an inbound call | Whether failover works. This is where the assumptions die |
| The 5pm test | Ring your own main number at 5:05pm, 7pm and Saturday morning | What a customer experiences outside hours, which is when a large share of regional enquiries arrive |
Do the fourth one
Nearly every business believes it has failover. A minority have tested it. The usual discoveries: the backup link was never provisioned, the failover was configured against a number that changed two years ago, or the whole arrangement depends on a specific person noticing and acting. Test it on a quiet Tuesday, deliberately, rather than discovering it during a storm. The related design questions are covered in business phone redundancy.
What to Watch Over the Next Twelve Months
Four dated things worth tracking, and one instruction on how to hold them.
The ACCC discussion paper
Due shortly after the inquiry's launch. It will define the questions being asked, which is the earliest real signal of where this is heading.
The regional forums
Being held across regional and remote Australia. If coverage genuinely constrains your business, this is a rare opportunity to put that on the record directly.
Optus satellite messaging
Indicated for 2026 subject to approvals. A second carrier changes the competitive picture and the handset compatibility picture.
The 2027 obligation's detail
The scope and definitions will determine whether it means anything for your site. “Outdoor” is doing a lot of work in the current framing.
The instruction: watch these, but do not wait for them. Every one of the five build-now decisions pays for itself regardless of what the ACCC decides, and none is made redundant by a roaming declaration. A business that spends the next twelve months getting reachable is in a better position under every possible outcome than one that spends it following a consultation.
If you are weighing up connectivity more broadly, how to read mobile coverage maps covers what the published categories actually promise, and the 6G roadmap covers the longer arc — including the lesson the 3G shutdown taught about relying on a layer that is scheduled to disappear.