Picture a cafe owner who switched off the surcharge on the terminals last night. This morning the first regular in the door asks, "So is the coffee dearer now?" That question is being asked all over Australia today. From 1 October 2026, businesses can no longer surcharge payments on eftpos, Mastercard or Visa, in store or online, and American Express has announced its own no-surcharge rules from the same date. For businesses that surcharged, the cost of accepting cards still has to be covered somewhere. This guide works through two illustrative examples, a cafe and a plumbing business that invoices, to show what the surcharge used to recover, what absorbing it costs over a year, how much a cheaper payment plan helps, and how small a price change usually needs to be. Then it covers the part most guides skip: making sure your phone greeting, on-hold message, AI phone agent and SMS all give customers the same clear answer.

Australia narrowly avoided a per capita recession in the June quarter, with real GDP growth of 0.4% doing little more than matching population growth, and on 29 September the Reserve Bank lifted the cash rate to 4.60%. Economists expect a slow finish to 2026. Every business owner is now deciding how to respond, and the instinct is to cut hard and wait. History suggests that is only half right. A Harvard Business Review study of 4,700 companies across three recessions found that 17% went bankrupt, were acquired or went private, and only about 9% came out stronger than before, and the difference was not how much they cut but what they cut. Australia's own downturns in 1991, 2008 and 2020 tell a similar story. This article looks at what the businesses that came out ahead did differently, why being easy to reach matters more when customers are ringing around, the mistakes to avoid, and a 30 day plan to put it into practice.

The Reserve Bank raised the cash rate to 4.60% on 29 September 2026, its fourth increase this year and a full percentage point since January. The Board was clear that it is not finished if inflation does not cooperate, saying it will act "including increasing the cash rate target further if needed". Most small business owners cannot do much about their loan rate beyond ringing the lender. What they can do is stop money leaking out through fixed costs that nobody checks, and the phone and internet bill is usually the worst offender. It is often split across three or four providers, it carries charges for staff who left and handsets bought years ago, and promotional prices end quietly. This guide sets out a 30 minute audit you can do this week, works through an illustrative eight person business that finds about $187 a month, lists the questions to put to your current provider, explains our half price nbn offer for new bundles, and flags the savings that end up costing more.

A physio clinic opens at nine. At 8.55 on a Monday the receptionist notices the phones are quiet, which is odd for a Monday, and then a patient walks in to say they have been getting a busy tone for twenty minutes. Somewhere between the clinic's internet, its phone provider and the platform that provider uses, something has gone wrong. What happens in the next hour depends almost entirely on one thing: who answers when the clinic calls for help. We walk through the same fault handled by three kinds of provider, a ticket only service, a reseller and a provider that runs its own network, to show why the experience differs so much. Then we set out what a tech partner who genuinely cares about your business does differently, how to interview a provider before you sign, the contract clauses to check, and where AI helps support rather than getting in the way.

When your ABN arrives, the internet hands you a checklist. Business name, GST, bank account, insurance, logo, website, socials, cards. The problem with a flat checklist is that it treats every item as equally important, and they are not. You can redesign a logo next month. You can rebuild a website in a weekend. But a few early decisions spread into places you do not control, and changing them later costs real money and lost customers. This guide sorts the usual post-ABN list by two simple questions: how hard is this to change later, and how much does it cost if I get it wrong? Move fast on the easy ones. Slow down on the few in the hard corner. Your business phone number sits right at the top of that corner, because within a year it lives in your customers' phones, your reviews, your listings and on the side of your van, and none of those copies can be updated.

A tenant rings a property management office at 10.40 on a Monday morning. The hot water system in their unit is leaking, there is water on the laundry floor, and they have a toddler. The property manager who answers is good at the job and she does what everyone does: she pulls up the tenancy, starts typing, asks the tenant to spell their surname, asks for the unit number again, types "HWS leaking", asks whether the water is hot or cold, types that, and somewhere in there misses the tenant saying the isolation valve is behind the washing machine and they cannot reach it. After the call she spends two minutes writing the note properly and booking a plumber, and during those two minutes two more calls go to the queue. Nothing about that call went badly. It is simply how phone work has always been done, and it costs more than it looks. In 2026 the phone system can write that note while the call is happening, with the valve detail in it, ready to check the moment the tenant hangs up. This article walks through what changes when it does: for the person on the call, for the customer waiting in the queue, and for the record. It also covers the habits that make AI notes trustworthy, the errors to check for, and the kind of AI note taking you should not allow anywhere near your customers.

Ask most business owners how old their phone system is and you get a shrug and a guess. Ask them what it costs each year and the guess gets vaguer. That is not carelessness. Phone systems are designed to be forgotten, and a good one is. The trouble starts when a system has been forgotten for so long that nobody notices it has stopped fitting the business. The receptionist keeps a notepad for messages because the voicemail is awkward. Two of the sales team give customers their mobile numbers because the desk phone cannot follow them. The after hours message still mentions a staff member who left in 2019. The maintenance invoice renews every year and nobody reads it. None of these is a fault, so none of them triggers a repair, and the system rolls on. This guide gives you a way to look at your own setup with fresh eyes in about twenty minutes. It is built around four simple tests: is the system still supported, is it secure, does it connect to how your people actually work now, and can it do what your callers expect. You walk the office, read the bill, ring yourself after hours, and write down what you find. By the end you will know whether to leave it alone, start planning, or move soon, and if it is time to move, you will know what to keep, what to retire and how to switch without losing a single call.

Picture the message as it landed. A work notification on a personal phone, apparently from the IT department: your password has expired, sign in here to reset it. Or, for some recipients, a note that their shift schedule had changed, with a link to check it. The link went to an address with the company name and the words "sso" or "okta" in it, and the page it opened looked exactly like the sign-in screen staff used every day. That is how the Twilio hack began in the first days of August 2022. A few employees typed in their username, password and one-time code, the attackers used those details on the real system within seconds, and from there they reached the internal tools Twilio staff use to support customers. Twilio detected the intrusion on 4 August, and the last unauthorised activity was on 9 August. By the time the investigation closed, 209 customer accounts had been accessed, 93 users of the Authy authenticator app had strange devices attached to their accounts, and Signal, the encrypted messaging app, had told around 1,900 of its users that their phone numbers or registration codes may have been exposed. There was no zero day exploit and no malware on a server. There was a text message, a copied page and a code that proved much less than everyone assumed. This article walks through the chain link by link, because each link has a direct equivalent in an Australian small business, and most of them can be fixed in a month.

Picture the call. Your phone buzzes with a sign-in approval you did not ask for, and a minute later somebody rings saying they are from support, that there is a problem with your account, and that they are sending a prompt to fix it. You are busy, the voice is calm and polite, and the prompts keep coming. That is more or less how Cisco was breached in May 2022, and the only unusual thing about it is that Cisco then published a detailed account of exactly how it happened. Two years later a different kind of problem hit Cisco's Webex platform: German journalists showed that meeting links for the Bundestag, ministries and the armed forces could be found by counting up and down, exposing the titles, times and attendee lists of thousands of meetings. The first incident came in through a person. The second came in through a product. Almost every communications security failure a business will face falls into one of those two buckets, and the fixes for both are cheaper than most people expect. This is a practical walk through both incidents, a thirty minute audit you can run yourself, and the questions worth asking whoever provides your phones and meetings.

Imagine the call. It is a Thursday afternoon and your office manager answers. The caller introduces themselves by name, says they are from your phone provider's security team, and mentions the incident that was in the news in July. They have your office manager's name, the business address and the main number, all correct. They explain that affected accounts need their admin access re-verified, and that a code will arrive by text in a moment. Could she read it back so they can confirm she is the account holder? Nothing about that call sounds wrong, which is the whole point, and it is the kind of call that becomes much easier to make after a breach like the one RingCentral disclosed on 28 July 2026. The data taken was contact information: names, emails, phone numbers and addresses for about 1.6 million records, verified independently in August. No passwords, no payment cards and no call recordings have been reported, and the platform itself was not affected. That makes it easy to dismiss. It should not be dismissed, because accurate contact details plus the name of the company to impersonate is exactly the combination a phone scammer needs, and the group responsible has spent two years showing that a well researched phone call gets further than almost any piece of malware.

Follow one customer through an ordinary week. On Tuesday they fill in the form on your website, which creates a lead in the CRM and sends somebody an email. On Wednesday they ring, because filling in a form and hearing nothing makes people ring, and whoever answers has no idea a form exists, takes the details again and writes them on a pad. On Thursday they text the mobile number from the van, which is a different number entirely and appears in no system at all. On Friday they message the business WhatsApp because the website has a button for it, and that message sits in an app on one person's phone. The following Tuesday they ring again and ask why nobody has got back to them, and the honest answer is that four different parts of your business each did roughly a quarter of the job and none of them knew about the others. Nobody was lazy. Every system worked exactly as designed. What failed is that there was no layer holding the thread, and adding AI to any one of those four systems would not have helped, because each of them would still have been reasoning about a quarter of the situation while sounding confident about all of it. That is the shape of the problem in most Australian businesses in 2026: plenty of intelligence, distributed across silos, with nothing in the middle. The fix does not involve replacing your CRM or your job management software, and it does not require a data project. It requires deciding which layer sits in the middle, and the answer is less obvious and more useful than most businesses expect.

The question gets asked backwards almost every time. People ask whether they should be using AI, which is a question about technology and therefore unanswerable without a sales pitch attached. The version that can actually be answered is narrower and duller: is there something in this business that happens often enough, similarly enough and expensively enough that a machine doing it would leave us better off after paying for the machine, the plumbing and the supervision. That has a number on the end of it, and the number is frequently small, which is a perfectly good outcome that nobody ever writes an article about. It is also sometimes very large, and the businesses where it is large tend to have one thing in common: they are leaking something they cannot see, usually calls that ring out, follow-ups that never happen, or notes that were never written. The leak does not appear in any report, because things that did not happen rarely do. So this is a worksheet rather than an argument. What the Australian evidence actually says once you understand why the adoption figures disagree so wildly, the four costs of which you will be quoted one, a payback calculation worked through three different shapes of business, the four situations where the answer should be no regardless of the arithmetic, and a thirty day sequence that costs almost nothing and ends with a number instead of an impression.

A business we spoke to earlier this year had an AI answering service running for five weeks and then turned it off. When we asked what went wrong the answer was interesting, because nothing did, exactly. It understood people fine. It booked appointments correctly. The problem was that nobody could say whether it was better than what came before, because what came before was a voicemail box that nobody had ever measured either. So when one customer complained about a call in week four, there was no counterweight. There were no transcripts anybody had read, no agreed number, no before and after, just one loud data point and a general unease, and the decision made itself. That story is far more common than the dramatic failures, and it is the reason a deployment plan is worth more than a better model. Gartner's estimate that more than forty per cent of agentic AI projects will be cancelled by the end of 2027 names escalating cost, unclear business value and inadequate risk controls, and the middle one is the quiet killer: not that the thing failed, but that nobody could demonstrate it had succeeded. The fix is boring and it works. Pick one narrow job. Write down what it may never do. Decide where a call goes when it stops. Write one sentence with a number in it that says what good looks like by day sixty. Then run it for a fortnight where no customer ever hears it, and read what it would have said. Everything after that is comparatively easy.

A customer rings on the twenty-ninth of December. They are not sure you are open, because it is that strange week where some businesses are trading normally, some are closed until the middle of January, and the shops are open but nothing else seems to be. Your phone rings out. There is no message, because whoever set the hours never thought about the twenty-ninth, which is not a public holiday and therefore behaves like an ordinary Tuesday as far as the system is concerned. So the caller hears eleven rings and nothing else, decides you are shut, and rings the next name on the list, who has a recorded message saying they are closed until the fourth but their on-call person can be reached by pressing one. The second business gets the job. Neither business will ever know that happened, because a call that rings out is not recorded as anything anywhere, and the only trace is a small unexplained difference in January revenue. That is the actual Christmas phone problem, and it has almost nothing to do with Christmas Day. It is about the four or five ambiguous days either side, the fact that public holidays and business closures are two different things that phone systems treat differently, and the reality that in most small businesses the person who understands the phone configuration finishes work on the nineteenth. All of it is preventable, none of it is difficult, and the work is much easier in October than in the week before.

There is a specific kind of blindness that affects every business about its own phone system. You cannot hear it. You know the options, you know which one you want, you know that operations handles bookings and that the second option is really for existing customers, and you have never in your life listened past the third option because you have never needed to. So when somebody suggests the menu might be a problem, it does not feel like one, and the conversation ends there. Meanwhile a caller who has never heard any of it is listening to forty seconds of talking, picking the option that sounds closest, waiting, reaching the wrong person, explaining the whole thing again, and quietly deciding something about your business that they will not tell you. The only way through that blindness is to stop looking at the configuration screen, which shows you what you intended, and start ringing your own number from a phone that is not part of your system, at times you would not normally ring, doing the things real callers do. Seven calls covers it. Most people find at least three things they did not know about, and roughly half find a path that rings forever and ends nowhere, which is the single most expensive fault in phone design and the one that never appears in a report. This is the audit, in order, with what each call is designed to expose, followed by the eleven fixes ranked by what they are costing you rather than by how easy they are.

The complaints arrive separately, over about two years, and each one gets handled on its own terms. A customer mentions in passing that they tried to ring last week and gave up. Someone in accounts says quietly that they seem to be answering a lot of calls that are not theirs. A new starter asks for everyone's mobile numbers because the directory is not much use. A manager notices that two people rang the same client on the same afternoon about the same thing. A long serving salesperson resigns and takes the clients with them, and nobody can quite explain how. Individually, each of these is a small operational annoyance with a plausible individual explanation. Together they are one thing: the phone system was designed for a building, everybody left the building, and the parts of it that depended on people being in the same room silently stopped working. That is not a criticism of anyone's decisions in 2020. The urgent problems then were laptops and video meetings and remote access, all of which got solved. The phones seemed fine because calls were still connecting, and the failures that followed are the kind that nobody escalates because each one looks like somebody's individual habit rather than a system fault. The useful thing about all six symptoms sharing a cause is that they also share a fix, it is mostly configuration rather than purchase, and it takes an afternoon.

The quote arrives per seat and the per seat number is three or four times what you pay now, multiplied by everybody in the business. It comes with a demonstration of a wallboard, a live queue, a supervisor listening in, and a set of graphs that make your current reporting look like a shoebox of receipts. None of that is dishonest and most of it works exactly as shown. The question nobody asks in that meeting is whether your business is shaped like the business the software was built for. Contact centre software solves one specific problem: a stream of interactions arriving faster than the people available to take them, needing to be distributed fairly and quickly across staff who are, for that call, interchangeable. If that is your Monday, the software is worth the money and doing without it costs you more in lost calls than the licences cost in dollars. If your callers ring a person they know by name, or if the phone rings four times an hour and everyone gets to it, then the wallboard will be a beautiful display of a problem you do not have. The frustrating part is that most businesses are neither. They have a queue in one part of the business, usually for about two hours a day, and named relationships everywhere else, and the arrangement that fits them is almost never the one they get quoted. This is the test, the six signs, the honest cost, and the split that actually works.

The demonstration will be good. That is the first thing to understand about buying AI phone answering in 2026: the part everybody watches has been solved for about two years, so every vendor can put a natural sounding voice in front of you that handles an interruption, copes with a broad Australian accent, and answers a question about your opening hours without sounding like a recording from 1998. If you judge on the demonstration you will judge them all identically, because on that measure they very nearly are identical. The differences sit entirely in the part nobody demonstrates: what the system is allowed to do, what it does when it does not know, where the call goes when it cannot help, and which country your customers' conversations end up sitting in. Those four things decide whether the product removes work from your business or simply moves the work to a quieter part of the day while adding a monthly bill. They are not technical questions and you do not need to understand how any of this is built to ask them. You need four sentences, asked in a particular order, and the discipline to watch the screen rather than listen to the voice. Here they are, with the answers that mean yes, the answers that mean no, and the answers that mean the person in front of you has not thought about it.

It happens without anybody deciding it. You start the business, you use your own mobile because that is what you have, and you ring customers from it because that is the phone in your hand. Later you get a proper business number (on the van, the website, the shirt) and you divert it to the same mobile, which feels like the moment you became a real business. But every call you make still goes out as your personal number, so customers keep saving that one, and within a year a solid share of your inbound calls arrive on a number that has no hours, no backup, no record and no way to be handed to anyone else. Then the compounding starts. The number rings at 6:40 on a Sunday morning. It rings while you are up a ladder. It rings when you are on holiday, and you answer, because the alternative is losing the job. If you take on an apprentice or an office person, they cannot help, because the business's phone is in your pocket and there is no way to share a pocket. And if it is a staff member's mobile rather than yours, then three years of customer relationships are stored under their name, on a handset that leaves when they do. None of this needs a new phone or a different carrier. It needs the business number to live in a system rather than on a handset, which takes about twenty minutes and turns on four settings almost nobody opens. This is that setup, and what happens if you skip each part.

Follow one business through four stages, because the argument is easier to see that way than in the abstract. Stage one: a sole trader gives out a personal mobile, answers when they can, and misses the rest. Stage two: they get a business number and divert it to the same mobile, which feels like progress and changes almost nothing, same phone, same missed calls, same personal number appearing on every outbound call to a customer who then rings it on a Sunday. Stage three: they hire someone, and now the number is the problem, because it lives in one person's pocket and there is no way for a second person to help. Stage four: they put in a virtual phone system, and for the first time the number belongs to the business rather than to a handset, it rings two people, knows what time it is, and when nobody can answer, something intelligent has a short conversation with the caller, captures what they want, answers the question they actually rang to ask, books them in, and leaves a summary that takes ten seconds to read between jobs. Nothing was installed. No cabling, no comms cupboard, no technician, and no handsets unless somebody wanted one. That is the shape of a virtual phone system, and for a business of one to twenty people it is now the most cost-effective communications arrangement available. Here is what it does, what it costs, and the three places it genuinely stops.

The conversation happens about four months in, usually when somebody in accounts flags it. The quote said $29 a user, there are twelve users, so the phone bill should be around $350, and it is $735. The immediate assumption is that someone has been slippery, and occasionally that is true, but far more often nothing improper has happened at all. What has happened is that a quote and an invoice are answers to two different questions. A quote prices seats: how much per person per month for the platform. An invoice prices a business: seats, plus the numbers you publish, plus the calls you make and the calls made to you, plus the AI minutes you used, plus the recordings you are storing for two years because your industry requires it, plus the promotional rate that ended in month thirteen. Every one of those is legitimate, disclosed somewhere, and absent from the one-page comparison that decided the purchase. The gap is typically a third to a half, and it is entirely predictable if you know where to look before signing rather than four months after. This article shows exactly where each part of the gap lives, with three worked monthly bills at four, twelve and forty users (quoted figure beside actual figure, line by line) then seven ways to close the gap honestly, and the six questions that make your quote and your first invoice agree.

The phrase "the new phone system is terrible" almost never describes a platform. Unpack it and you find four things, in roughly this order: calls reach the wrong person, nobody is confident transferring, the after-hours message is wrong or missing, and about a third of the team never properly set up the mobile app so calls appear not to ring. All four are sequencing failures. All four were decided by what got done last rather than by what got decided first. And all four are cheap to prevent and expensive to fix, because once a business is limping along on a phone system it dislikes, nobody volunteers to reopen it, so a two-week problem becomes a three-year one. This article lays the same work out as a fourteen-day plan with a name against every line. It is deliberately boring: dates, owners, and a test script. The interesting judgement all happens on day one and day two, before anything is ordered, and the rest is execution. If you already have a provider and a signed quote, you can start at day three. If you are still comparing quotes, do days one and two anyway: the answers will change which quote you accept, and they are the only part of this nobody else can do for you.

The demo is at eleven on a Tuesday. A salesperson on a good connection, in a quiet room, speaks clearly to their own AI receptionist. It answers warmly, understands immediately, books an appointment and sends a confirmation. It is genuinely impressive, and it is also completely uninformative, because every provider's demo goes exactly like that, including the ones whose product will frustrate your customers within a fortnight. The reason is simple. A demo is a rehearsed conversation between two cooperating parties, and a business phone call is nothing of the sort. Your callers interrupt. They mumble a suburb name from a moving car. They start explaining before the greeting has finished, change their mind halfway, ask for a person, ask something you have never published, and occasionally ring at ten past nine at night because something has gone wrong. None of those appear in a demo and all of them appear on Tuesday. This article is the fix, and it costs one afternoon: ten specific calls, made by you, to each shortlisted provider's trial number, in the same week, from the same phones. Ten calls is enough. In every comparison we have watched a customer run, the ten calls sorted a four-way tie into a clear first and a clear last, and the deciding failures were never the things anybody had been comparing.

A practice manager forwarded us an email in the middle of 2026 with the subject line "An exciting update about your service" and asked, reasonably, whether it mattered. It said the business was being transitioned to a new platform, that the plan and pricing were unchanged, and that no action was required. It was, as far as we could tell, entirely honest. It also meant that in about six weeks the practice's call recordings would live on a system that did not import them, its after-hours on-call routing, built up over four years and understood by nobody who currently worked there, would need rebuilding from memory, its two most-advertised numbers were registered to a company that had been deregistered in 2021, and its contract would auto-renew for thirty-six months eleven weeks later. None of that was hidden. It simply was not in the email, because the email was about the migration and those were consequences of it. This is the ordinary shape of provider change in Australia right now, after a year in which customer bases changed hands, enterprise assets moved between carriers, small-business voice platforms were rebuilt on partners' technology, and legacy products were given end dates. This article is about reading that email properly: which of four quite different events it describes, what the four require, and a two-week response that costs an afternoon and removes almost all of the risk.