The spreadsheet beside the system is the system telling you something.
You have outgrown your software when a spreadsheet lives beside it, the plan cap runs your workflow, the API sits on a tier you do not pay for, the export is incomplete, nobody can define a field, a workaround costs more than the fix, or the bill outgrows revenue. Most businesses have outgrown the joins, not the software.
Sign one: a spreadsheet lives beside the system
The spreadsheet exists because the software could not do a thing, and somebody had to. It is the most honest documentation you own, and it is usually where the real process lives. One spreadsheet is a workaround. Three is a second system nobody is maintaining.
This is the common condition, not the exception. On the ABS digital intensity index, 59% of businesses with up to four staff sat in the lowest “baseline” band and 1% of all businesses reached “advanced” (ABS, Development of a Composite Indicator for Business Digital Intensity in Australia, 2021–22 data, article released 28 July 2023; the index is four years old, and we say so rather than implying it is current). Baseline usually means the software is a ledger and the spreadsheets are the business.
The test: if the spreadsheet vanished tonight, could you run tomorrow? If not, you have outgrown either the software or the way it is wired to everything else. The difference matters, and it is the last section of this page.
Sign two: the plan cap is running your workflow
Every product has a number where the price steps. ServiceM8 steps at 30, 50, 150, 500 and 1,500 jobs a month, from free to $349 (ServiceM8 pricing, checked September 2026). MYOB Business Lite connects at most two bank accounts and runs payroll for at most two employees (MYOB pricing, checked September 2026). QuickBooks Online allows 1, 3, 5 and 25 users across its four plans (QuickBooks Online AU pricing, checked September 2026). Xero’s Grow, Comprehensive and Ultimate plans run payroll for 2, 5 and 10 people at $78, $107 and $143 a month including GST (Xero AU pricing, checked September 2026).
None of that is a problem until you are batching jobs into next month to stay under the cap, sharing one login between three people, or paying a fourth employee outside payroll because the plan only does two. When the cap decides how you work rather than how much you pay, you have outgrown the tier — and possibly the product.
Sometimes the honest answer is simply the next tier. Check the price of the step before concluding you need something else; it is usually the cheapest fix on this page.
Sign three: the API sits on a tier you do not pay for
The moment you want two systems to talk, the API becomes the product. ServiceM8 lists API access on Premium at $149 a month and above; the tiers below do not list it (ServiceM8 pricing, checked September 2026). Simpro publishes no pricing at all, so the API question is part of a sales conversation (Simpro pricing, checked September 2026).
If the automation you want carries a tier jump as a hidden prerequisite, the software is telling you what it thinks you are: a customer of the smaller kind. That is not a moral judgement, but it is a real cost, and it recurs every month for as long as the connection exists. The feasibility checklist starts with exactly this question because it is the one that most often turns a cheap product into a dear one.
Sign four: the export is not complete
You are required to keep most business records for five years and to be able to extract digital records into a standard format such as Excel or CSV (ATO, record-keeping rules for business, checked September 2026). If your software exports the ledger but not the attachments, the notes, the job history or the custom fields, it has your data and you have a report.
Test it before you need it. Export everything, open the files, count the rows. Then check how much a move would actually carry: the standard Xero conversion brings across the current and previous financial year, with each additional year $150 ex GST (Jet Convert AU pricing, checked September 2026, ex GST). A product you cannot fully leave is a product that has outgrown you, not the other way round — and the migration checklist begins with this test for that reason.
Sign five: nobody can say what a field means
“Status” has eleven values and three people use it three ways. “Customer” sometimes means the site and sometimes the account. A date field holds the date somebody remembered to fill it in. When the data no longer describes the business, every report is an argument and every automation built on it inherits the argument.
This is a human problem the software made visible, and automating it makes it worse faster. Human error accounted for 37% of data breaches notified to the OAIC in the first half of 2025 (OAIC, Notifiable Data Breaches statistics January–June 2025, published 4 November 2025); ambiguous fields are how a wrong record reaches a wrong customer. The fix is a one-page data dictionary — what each field means and who is allowed to change it — written before any new system is chosen. Half the time, once it exists, the current software turns out to be fine.
Sign six: the workaround costs more than the fix
Price the workaround honestly. Full-time average weekly earnings were $2,083.70 in May 2026 (ABS, Average Weekly Earnings, Australia, May 2026, released 13 August 2026); with 12% super (ATO, Super guarantee, rate from 1 July 2025) that is about $62 an hour loaded over paid hours. An administrator on the Clerks Award Level 2, at $29.45 an hour from 1 July 2026 (Fair Work Ombudsman, Clerks — Private Sector Award pay guide, rates from 1 July 2026), loads to roughly $33. Six hours a week of re-keying is $19,400 a year at the first rate and $10,400 at the second.
Compare that with the fix. If the fix is a tier upgrade at $70 a month, the workaround lost the argument years ago. If it is a purpose-built connection between the two systems, run the payback arithmetic on the ROI page, including the halved-hours case. If it is replacing the whole system, add the migration cost, the training and the integrations you will break, then compare again. Replacement wins less often than people expect at this stage, because by now they are tired of the workaround.
Sign seven: the bill grows faster than the business
Per-seat and per-task pricing rise with headcount and activity, whether or not the software does more for you. Zapier bills per completed task from 100 a month free through paid tiers starting at US$19.99 (Zapier pricing, checked September 2026, USD); every automation that works costs more than the one that did not. Award wages rose 4.75% from 1 July 2026 (Fair Work Ombudsman, Annual Wage Review 2026, from 1 July 2026); if your software bill rose faster than your wage bill and your revenue, the software is taking a growing share of a business it did not grow.
It is worth separating the bill from the value. Over one third of businesses reported a skills shortage in 2024–25 and 15% named lack of funds as their top barrier to innovating (ABS, Characteristics of Australian Business, 2024–25, released 25 June 2026). A subscription that removes a hire is cheap at any price. A subscription that scales with a hire you made anyway is a tax.
Two signs that are not signs
“It looks dated.” Ugly software that does the job and exports cleanly is better than handsome software that does not. “A competitor uses something newer.” Their process is not yours. And the loudest non-sign this year: “it doesn’t have AI.” Just 12% of Australian businesses reported using AI in 2024–25, up from 1% two years earlier (ABS, Characteristics of Australian Business, 2024–25, released 25 June 2026); a separate Commonwealth tracker puts small business adoption around three times higher on a different definition (National AI Centre, AI Adoption Tracker, Department of Industry, Science and Resources, January – March 2025). Either way, the businesses we talk to are not losing money for want of AI. They are losing it to re-keying.
Most of what we say about AI tools comes back to that: fix the joins first, because a model bolted onto bad data produces confident nonsense. Software you have genuinely outgrown shows up in the seven signs above, not in the version number.
- KNDR·01Spreadsheet beside the systemSign. Usually a joins problem.
- KNDR·02Plan cap runs the workflowSign. Check the next tier first.
- KNDR·03API on a tier you do not pay forSign. Price the tier jump into every automation.
- KNDR·04Incomplete exportSign. Test it before you need it.
- KNDR·05Undefined fieldsSign. Fix with a data dictionary before any replacement.
- KNDR·06Workaround dearer than the fixSign. Run the arithmetic, including the halved case.
- KNDR·07Bill outgrowing revenueSign. Separate the bill from the value it buys.
- KNDR·08Looks dated · competitor uses X · no AINot signs. Ignore them until the seven above are clear.
Outgrown the software, or outgrown the joins
Most of the time it is the joins. Your accounting package is fine. Your job management system is fine. They do not talk, so somebody re-types and the spreadsheet grows. Building that one connection costs a fraction of replacing either system, removes the actual pain, and is reversible if you change products in two years — it is most of what we build, and we say so on the page that sells custom software.
Replace the system when the caps, the export and the API tier all fail together, because then the vendor has decided you are not their customer. Connect the systems when the failures are between them. The scorecard below sorts one from the other.
Ten questions about your main system. The result separates three answers: keep it, connect it, or replace it.
- Is there a spreadsheet you could not run tomorrow without?
- Are you changing how you work to stay under a plan cap?
- Does an automation you want require a tier jump for API access?
- Have you tested the full export, and did it miss attachments, notes or history?“Yes” means it is incomplete.
- Do three people use the same field three different ways?
- Have you priced the workaround at a loaded hourly rate, and did it exceed the fix?
- Has the software bill grown faster than revenue over two years?
- Does the vendor publish its prices and its API limits?“No” counts toward outgrown.
- Is the main pain between systems rather than inside one?
- Could a competent developer connect the two with the documentation that exists?
0 of 10 answered. The result appears once you have answered them all.
Runs in your browser; nothing is sent. If it says the joins, that is the answer we give on most first calls, and it is the cheaper one.
The questions people actually ask.
01Our vendor says the next tier fixes everything. Should we believe them?
Check the three things a tier cannot fix: the export, the field definitions and the connections to other systems. If those are the pain, the tier moves the cap and leaves the problem. If the pain is genuinely the cap, the tier is the cheapest answer on this page and you should take it.
02We have outgrown it but a migration terrifies us. What then?
Connect first, migrate later. A connection between the old system and whatever it is failing to talk to buys you a year and is not wasted when you move — you rebuild one connector, not a platform. Then run the migration on your timetable rather than the vendor’s.
03Isn’t “build the joins” just what you sell?
Yes, and it is also the cheaper answer more often than replacement, which is why we sell it. When the caps, export and API tier fail together we say replace — and the ledger conversion itself is a job for a conversion partner, not a custom build. Xero’s does it for less than we could.
Stop bending.
Start shipping.
Bring the one flow that keeps costing you a Tuesday afternoon. We map it live and tell you what we'd build, what we wouldn't, and what it costs.