The Ninety-Second Test
Take a blank page and write down every company that invoices your business for something technological. Do not look anything up — the value of the exercise is in what you cannot recall, not in what you can.
- Write the list from memory. Give yourself ninety seconds and stop.
- Now open the last three months of the business card statement and add what you missed.
- Circle everything on the list that is a monthly or annual recurring charge.
- Underline everything you cannot immediately explain the purpose of.
Two things reliably happen. The list from the card statement is longer than the list from memory, usually by three or four entries. And there is at least one underlined item — a subscription nobody can account for, still charging, months or years after the person who signed up for it stopped using it or left.
This is not a story about disorganised businesses. It is what happens to every business that grows, because the alternative — appointing somebody to own the whole technology estate and design it coherently — requires a role that a twenty-person company cannot justify and a two-hundred-person company only just can. The fragmentation is structural. So is the fix, which is why it works.
Why Every Estate Ends Up This Way
Trace any fragmented estate backwards and the pattern is always the same. Each purchase was rational, urgent, and made by whoever was standing closest to the problem.
| What got bought | What triggered it | Who chose it |
|---|---|---|
| Phone system | The old one failed, or the provider announced it was ending the product | Whoever answered the letter |
| Internet | A move, or an outage that lasted long enough to be intolerable | Whoever was in the office that week |
| Cameras | A break-in, or a near miss at a neighbouring business | The owner, on a Saturday |
| Scheduling or field tool | A dispute about who was where | The operations manager |
| Video conferencing | 2020 | Everybody, separately |
| SMS or notification service | Too many missed appointments | Reception |
| E-signature, forms, storage | A single awkward transaction | Whoever was embarrassed by it |
Notice what is absent from that table: any point at which somebody asked whether the business already had the capability. That question requires a person who knows the whole estate, and in a fragmented estate no such person exists. The fragmentation is self-reinforcing — the more suppliers you have, the less likely anyone is to know that the new tool duplicates an old one.
The Six Costs With No Budget Line
Every one of these is real, recurring, and invisible to a profit-and-loss statement organised by supplier.
1. Subscriptions on hardware you own
The device was a capital purchase. The features that make it useful are a monthly fee, per device or per site, forever. Over five years this frequently exceeds what the hardware cost.
2. Capability paid for twice
Conferencing, SMS, call recording, an answering service, a scheduling calendar — bought standalone from a platform that already includes them. This is usually the single largest recoverable line.
3. App and login sprawl
Every app is a login, a password reset path, an offboarding step and a security surface. Nine apps is nine places a departed employee might still have access.
4. Induction time, every hire
Somebody teaches each new person the estate. Two systems is an afternoon. Nine systems with nine notification models is a running cost that recurs with every hire and never appears in a budget.
5. Nine warranty and support processes
Different portals, different hours, different escalation paths, different definitions of urgent. Under pressure, the time goes on finding the right process rather than fixing the fault.
6. The blame gap
When two suppliers each say the fault is the other's, the business pays for the argument in downtime and in somebody's afternoon. This one is covered properly further down, because it is the worst of the six.
Subscription Creep on Hardware You Own
This deserves its own section because it is the fastest-growing line in most small business technology spend and the least examined.
The model is simple and it works: price the hardware attractively, then charge monthly for the part that makes it useful. Cloud recording. Retention beyond a few hours. Detection features. Sometimes remote access at all. Stop paying and the device does not stop existing — it stops doing the thing you bought it for. That is a rental arrangement wearing the clothes of a purchase, and businesses sign up to it one device at a time without ever seeing the aggregate.
The test that reveals it
For every connected device in the business, ask one question: what stops working if I stop paying? If the answer is "nothing", you own it. If the answer is "the recording, the history, or the ability to see it remotely", you are renting it, and the rent has no end date. Neither answer is automatically wrong — but you should know which one you are in, and most businesses do not.
The same question applied to software is equally useful, and it is why published, per-seat pricing beats bespoke pricing over any reasonable horizon. A published rate is a number you can check at renewal. A privately negotiated rate is a number with nothing to benchmark it against, which feels like a win on signing day and is a problem two years later when you have no idea whether you are being looked after or quietly repriced.
How to Audit It Properly
Most estate audits fail for one reason: they are done from the invoices the bookkeeper files, and the expensive items are not there. They are on a card, authorised by email, renewing silently.
Build a five-column table. Nothing more sophisticated is required and anything more sophisticated will not get finished.
| Column | What goes in it | Why it matters |
|---|---|---|
| Supplier | Who charges you | The count itself is the headline finding |
| What it does | One sentence, in plain words | Duplicates become visible the moment two rows say the same thing |
| Monthly cost | Normalise annual charges to monthly | Annual billing is where the forgotten subscriptions hide |
| Seats or devices | How many, and how many are actually in use | Unused seats are typically ten to twenty per cent of a per-user bill |
| Renewal and exit | Renewal date, notice period, exit cost | This column is where the real money is decided and it is the one always left blank |
Three sources, in this order
Card statements first — twelve months, not three, so annual renewals appear. Then the app stores and browser password manager, which reveal services nobody is invoicing to the business at all. Then the invoice file, which is the tidiest source and the least complete. Doing it in the reverse order is why most audits find nothing.
The Sixty-Month Number
Monthly figures do not change behaviour. A business will look at $89 a month for years without blinking. The number that changes behaviour is the same figure over the life of the decision.
×12
Normalise everything to annual
×5
Five years is a realistic hold period
10–20%
Typical share of per-seat spend on unused seats
3–4
Suppliers a typical owner cannot name from memory
Total your recurring column, multiply by sixty, and put that figure at the top of the page. We are deliberately not inventing a dollar total for a hypothetical business, because yours will be different and a fabricated benchmark is worse than no benchmark. What we can tell you is the shape of the answer: it is larger than people expect, the duplicated capability line is usually the biggest single item, and the unused seats line is the easiest to recover because it requires no negotiation with anybody.
The Order to Fix It In
Sequence matters more than effort here. Businesses routinely start at step four, get a small win, and never reach the steps where the money was.
- Cancel the duplicates. Anything your existing platform already includes. No negotiation, no migration, no permission required. This is free money and it is usually the largest line.
- Remove the unused seats. Licences for departed staff, duplicate accounts, extensions on desks nobody sits at. Also free, also nobody's job, which is why it never gets done.
- Read the contract terms. Renewal dates, notice periods, exit costs, and the price review clause almost nobody has read. You cannot negotiate from a position you have not established.
- Consolidate what genuinely overlaps. Now — after you know what you actually use and what you are contractually able to move. Not before.
- Then talk about price. Last, because a discount on a service you should have cancelled is the least valuable outcome available, and you only get one moment of leverage per term.
Why steps one and two come first
They require nobody's approval, carry no migration risk, and can be completed in an afternoon. They also change the conversation in step five: a business that has already removed its waste negotiates from a clean position, while a business that has not is asking for a discount on padding. Do the free things before the hard things.
The Blame Gap, and What It Costs Per Incident
Of the six costs, this is the one that turns a bad morning into a bad week.
The scenario is familiar to anybody who has run a site. Calls are dropping. The phone platform vendor runs a test, finds nothing wrong on their side, and points at the internet. The internet provider runs a line test, finds it within specification, and points at the phone platform. Both are being honest. Both are also finished, because each has confirmed their own scope and neither owns the gap between them.
The most expensive hour in a fragmented estate is not on anybody's invoice. It is the hour somebody in your business spends proving to two suppliers that one of them is wrong.
— the argument you pay for either way
The gap is real and it is where genuinely hard faults live: asymmetric upload contention that only appears at 4pm, a router doing something clever with packet prioritisation, a handover between two networks where neither party can see both ends. Nobody diagnoses that without visibility on both sides of it.
This is the single strongest operational argument for consolidation, and it is stronger than the financial one. When the network and the platform are the same provider, there is no gap to fall into and no argument to have. The fault is ours by definition, which changes the incentive from proving innocence to fixing the problem.
What Consolidating With Us Actually Covers
Being specific here is more useful than being expansive, so this is the honest scope rather than an aspirational one.
The platform
Cloud phone system with desktop and mobile apps, AI phone agents with natural Australian accents answering around the clock, untimed HD video meetings for up to thirty participants, two-way business SMS, call recording, IVR, call queues and ring groups. Most of the duplicated capability in a typical estate is in this list.
The network
We own and operate our own network rather than reselling somebody else's, with 99.99% uptime and Australian hosting. We can also supply the internet — nbn plans alongside the phone service — which is what closes the blame gap described above.
The hardware
Desk and cordless business handsets and headsets, and support for standards-compliant certified IP handsets you already own. App-first by design, so hardware is a choice rather than a requirement.
The connections
Integrations with Salesforce, HubSpot, Zoho, Xero, Monday and over a thousand other applications, plus open APIs. This matters for consolidation because it means the specialist tools you keep still talk to the platform.
What it does not cover: your accounting package, your point of sale, your industry-specific field software, your cameras or your document storage. Those stay where they are, and the integration layer is how they connect rather than something we replace. Any supplier claiming to consolidate all nine is selling you a migration you will regret.
Two Cases Where Consolidating Is Wrong
1. A working system, mid-contract
If a system works, staff know it, and there are eighteen months left on the term, moving it costs an exit fee, a migration risk and a retraining exercise to buy a tidier diagram. Consolidate at natural replacement points — renewal, a failure, a new site, a price rise. Not on a schedule that suits a diagram.
2. A genuine specialist tool
Industry-specific software usually beats a general-purpose equivalent at the thing it was built for, and the switching cost is measured in workflow rather than dollars. The right move is to integrate it, not replace it. Consolidation is about removing redundancy, not about reducing a vendor count for its own sake.
There is a third caution worth stating, because it applies to us as much as to anyone. Consolidating creates a single point of commercial dependency. That is a genuine trade-off and pretending otherwise would be dishonest. It is manageable — but only if you check the exit before you commit.
Test the Consolidated Supplier for Lock-In
Four questions. Ask them of us and of every alternative, and compare the answers rather than the brochures.
| Ask | The answer that means it | The answer that does not |
|---|---|---|
| Do I keep my numbers if I leave? | Yes, they port, and here is who holds the rights of use | Numbers are provided as part of the service |
| Is your pricing published? | A public price list you can read before you call | Bespoke, tailored, individually quoted |
| Can my software get at my data? | Open APIs, documented, with real integrations in production | An integrations marketplace with no API behind it |
| Is the hardware standards-based? | Standard SIP handsets, releasable if you go | Proprietary devices that only work here |
Our answers are yes, yes, yes and yes, and we would rather you tested them than believed them. A provider that has removed its own lock-in mechanisms is a provider that has to keep earning the relationship, which is the arrangement you want to be in.
A Two-Week Plan
- Day one. The ninety-second test, then twelve months of card statements. Build the five-column table. Leave the renewal column blank for now.
- Days two to four. Fill in the renewal and exit column. This means finding contracts, which is the tedious part and the part that pays. Anything you cannot find, request in writing from the supplier.
- Day five. Cancel the duplicates and the unused seats. Do not wait for the rest of the plan — these are independent wins.
- Week two. Rank what remains by monthly cost, then decide which items are genuinely overlapping, which are specialist keepers, and which are simply mid-contract and should wait.
- End of week two. One conversation with one consolidated provider, holding a table rather than a vague sense of frustration. That table is the whole difference in how the conversation goes.
The summary
Nine suppliers is normal, not negligent. The costs that hurt are subscriptions on hardware you own, capability bought twice, and the blame gap between two vendors who are each individually correct. Audit from the card statement, project it over sixty months, and fix it in order: duplicates, unused seats, contract terms, consolidation, price. Leave working systems and genuine specialist tools alone. And before you consolidate onto anyone, including us, ask what leaving looks like — the answer tells you everything about how the next five years will go.
Related reading: the same argument run over communications spend, what one platform actually replaces, when bundling phone and internet is right and when it is not, and the devices and apps that make up our side of the estate.
Frequently Asked Questions
How do I find every technology subscription my business is paying for?
Work from three sources in a specific order, because the order is why most audits find nothing. Start with twelve months of business card statements rather than three — annual renewals only appear over a full year, and card charges are precisely where the forgotten subscriptions live, since they were authorised by email by whoever needed the tool and never entered a formal procurement process. Second, check the app stores on company devices and the browser password manager, which reveal services nobody is invoicing to the business at all, often signed up with a personal card and reimbursed once. Only third go to the invoice file your bookkeeper maintains, which is the tidiest source and the least complete. Then build a five-column table: supplier, what it does in one plain sentence, monthly cost with annual charges normalised to monthly, seats or devices with a note on how many are genuinely in use, and renewal date with notice period and exit cost. The second column is what makes duplicates visible, because two rows describing the same capability in plain words is impossible to miss. The fifth column is where the real money is decided and it is the one always left blank.
What is subscription creep and how do I tell if I am affected?
Subscription creep is the accumulation of recurring fees attached to hardware you have already bought outright, and it has become the fastest-growing line in most small business technology spend. The commercial model is straightforward: price the device attractively, then charge monthly for the capability that makes it useful — cloud recording, retention beyond a few hours, detection features, sometimes remote access at all. Stop paying and the device does not disappear, it simply stops doing the thing you bought it for, which makes it a rental arrangement wearing the clothes of a purchase. Businesses sign up one device at a time and never see the aggregate. The diagnostic question for every connected device you own is: what stops working if I stop paying? If the answer is nothing, you own it. If the answer is the recording, the history, or the ability to view it remotely, you are renting it and the rent has no end date. Neither answer is automatically wrong — sometimes a service genuinely is being delivered — but you should know which situation you are in, and most businesses have never asked. Apply the same question to software, then total every recurring charge and multiply by sixty months. That figure, not the monthly one, is what changes decisions.
In what order should I consolidate suppliers to save the most money?
Duplicates, unused seats, contract terms, consolidation, then price — and the sequence matters far more than the effort you put in, because most businesses start at the last step, secure a small discount and never reach the steps where the money actually was. First cancel anything your existing platform already includes: conferencing, SMS, call recording, an answering service, a scheduling calendar bought standalone alongside a platform that bundles them. This needs no negotiation, no migration and nobody's permission, and it is usually the single largest recoverable line. Second, remove unused seats — licences for departed staff, duplicate accounts, extensions on desks nobody sits at — which typically account for ten to twenty per cent of a per-user bill and which providers rarely volunteer. Third, establish your contract position: renewal dates, notice periods, exit costs and the price review clause almost nobody has read, because you cannot negotiate from a position you have not established. Fourth, consolidate what genuinely overlaps, now that you know what you use and what you are contractually free to move. Only then discuss price. A discount on a service you should have cancelled is the least valuable outcome available, and you get one real moment of leverage per contract term.
Why does having multiple suppliers make faults take longer to fix?
Because of the blame gap, which is the most expensive of the fragmentation costs and the one least visible on any invoice. The scenario runs the same way every time: calls start dropping, the phone platform vendor runs their tests, finds nothing wrong within their scope and points at the internet connection; the internet provider runs a line test, finds the service within specification and points back at the phone platform. Both suppliers are being honest and both have finished work, because each has confirmed their own scope and neither owns the space between them. Meanwhile somebody in your business spends an afternoon relaying test results between two companies in order to prove that one of them is wrong, and the calls are still dropping. The faults that live in that gap are the genuinely hard ones — upload contention that only bites at four in the afternoon, a router prioritising packets in an unhelpful way, a handover between two networks where neither party can see both ends — and nobody diagnoses those without visibility on both sides. When the network and the platform belong to the same provider there is no gap to fall into and no argument to have, which changes the incentive from proving innocence to fixing the problem. That is a stronger argument for consolidation than the financial one.
When is consolidating suppliers the wrong decision?
In two clear cases, and it is worth naming them because a provider who claims consolidation is always right is selling rather than advising. The first is a working system mid-contract. If it does the job, your staff know it, and there are eighteen months left on the term, then moving it costs an exit fee, a migration risk and a retraining exercise, and buys you a tidier diagram. Consolidate at natural replacement points instead — a renewal, a failure, a new site, a price rise, the end of a commitment — rather than on a schedule that suits an org chart. The second is a genuine specialist tool. Industry-specific software usually beats a general-purpose equivalent at the specific thing it was built for, and the switching cost is measured in workflow disruption rather than in dollars. The correct move there is to integrate it, not replace it, which is what open APIs and a large integration catalogue are for. Consolidation is about removing redundancy, not about reducing a vendor count for its own sake. There is also a third caution that applies to every consolidated supplier including us: it concentrates commercial dependency in one relationship, which is a real trade-off you should manage by checking the exit terms before you commit rather than by pretending the trade-off does not exist.
How do I check whether a consolidated provider will lock me in?
Ask four questions and compare the answers rather than the brochures. Do I keep my numbers if I leave? The answer that means it is yes, they port, and here is who holds the rights of use in writing — the answer that does not is that numbers are provided as part of the service. Is your pricing published? A public price list you can read before you call gives you a benchmark at renewal; bespoke, tailored or individually quoted pricing feels like special treatment on signing day and removes the only reference point you will want two years later. Can my own software get at my data? Documented open APIs with real integrations running in production is a genuine answer; an integrations marketplace with no API behind it is a screenshot. Is the hardware standards-based? Standard SIP handsets that can be released if you go are portable; proprietary devices that only function on one platform are a deposit you forfeit. Our answers are yes to all four, and we would much rather you tested them than believed them, because a provider that has deliberately removed its own lock-in mechanisms has to keep earning the relationship every year — which is exactly the arrangement a customer should want to be in.
What should a consolidated provider actually cover, and what should stay separate?
A realistic consolidation covers the communications platform, the network underneath it, the endpoint hardware and the integration layer — and explicitly does not cover your specialist business software. On our side that means the cloud phone system with desktop and mobile apps, AI phone agents answering around the clock with natural Australian accents, untimed HD video meetings for up to thirty participants, two-way business SMS, call recording, IVR, call queues and ring groups; the network itself, which we own and operate rather than resell, hosted in Australia with 99.99% uptime, including nbn plans so the internet and the phone service come from the same accountable party; desk and cordless handsets and headsets plus support for standards-compliant IP handsets you already own; and integrations with Salesforce, HubSpot, Zoho, Xero, Monday and over a thousand other applications alongside open APIs. Most of the capability duplicated in a typical fragmented estate is somewhere in that first list, which is why the duplicate-cancellation step is usually the largest saving. What stays separate is your accounting package, your point of sale, your industry field software, your cameras and your document storage. Those connect through the integration layer rather than being replaced, and any supplier claiming to absorb all nine of your vendors is proposing a migration you will regret.
