If you employ people on a flat loaded rate in Australia — one clean hourly figure instead of the award's lattice of penalties, overtime tiers and loadings — the award has not gone away. It still applies in full, and your flat rate only satisfies it through a mechanism that carries a proof burden: a common-law set-off clause (which works only to the extent that, over the period, total pay actually covers total award entitlements — and the employer carries the reconciliation), an award annualised-wage arrangement (which mandates a periodic reconciliation outright), or an individual flexibility arrangement (which the Act requires to leave the employee better off overall). Three legal doors, one identical piece of arithmetic behind all of them: award-equivalent pay versus actual pay, per person, over real hours.
Which is why almost every employer and advisor who runs that arithmetic calls it the same thing: a BOOT test. Everyone in payroll knows the artefact: a workbook with the Fair Work pay guide pasted into one tab, four weeks of hours typed into another, and a formula column that ends in PASS.
A note on the name
Strictly, the BOOT — the Better Off Overall Test, s.193 of the Fair Work Act — is the Fair Work Commission's test for approving an enterprise agreement, and it lives (and keeps applying) for the life of that agreement. We've written about that discipline separately in The BOOT Doesn't Stop at Approval. What loaded-rate award employers run is a better-off-overall reconciliation in the BOOT's image — the industry borrowed the name because the arithmetic is the same. We use the borrowed name in this article because it's the one on the spreadsheets our clients actually maintain; the legal mechanism underneath is the set-off / annualised-wage / IFA obligation described above. Same engine, two legal frames.
Recently a client shared theirs with us — a genuinely careful one. Four weeks, one full-time field worker, every day decomposed into ordinary time, overtime tiers, allowances, leave loading. It passed, all four weeks. It was better than most we have seen. And when we replayed it against the systems of record, three things surfaced that no diligence inside a spreadsheet could have prevented.
What reconciling one careful manual BOOT test surfaced — in both directions
The two records disagreed on three of nineteen days — and nobody had known. On one day the workbook understated the hours by more than three hours against the time-and-attendance record. On another, the time system itself turned out to hold a duplicate card — the workbook was right, and the duplicate had sat unnoticed. And one worked public holiday appeared in the workbook with no trace in the time system at all. Until the two records were compared, there was no way to know which to trust.
The award rate matched no year. The benchmark hourly rate sat roughly 3% above the rate actually in force for the test period — pasted from a pay guide of uncertain vintage. In the client's favour, as it happened: the error made their test harder to pass, and their results stood. But an unanchored rate is an unanchored process — Fair Work moves the floor every year, from the first full pay period on or after 1 July, and a pasted guide doesn't.
It was a snapshot. One employee, one four-week window. The method has no mechanism to re-run itself when the floor moves, rosters shift, or weekend patterns change — precisely the events that move a better-off-overall margin.
None of that is carelessness — on either side. It is what happens when two records that are supposed to describe the same hours are never systematically compared. Hand transcription, static rates and point-in-time sampling are structural properties of doing this in a spreadsheet — and until recently, they were tolerable ones.
Why "tolerable" ended
Three things have moved underneath this ritual.
Underpayment became a crime. From 1 January 2025, intentionally underpaying wages or certain entitlements is a criminal offence under the Fair Work Act, carrying up to ten years' imprisonment. Be precise about what that does and doesn't mean: the offence turns on intent, and the Fair Work Ombudsman is explicit that honest mistakes are not caught by it. Where your systems do bear on your exposure is one layer down — in whether a contravention was knowing or reckless, which lifts it into the serious-contravention penalties, and in the Voluntary Small Business Wage Compliance Code, where a small business that can show it took the steps in the Code cannot be referred for criminal prosecution. In both, the evidence is the same: what were you doing to know?
Casual employment turned on substance. Since August 2024 the Act defines a casual by the real substance of the working relationship rather than the label on the contract, and eligible casuals can give notice under the employee choice pathway to move to permanent employment. Whether your loaded rates actually clear the award for people working firm, regular patterns is a live question every pay cycle, not a hire-date formality.
The floor moves every year. Annual wage review increases apply from the first full pay period starting on or after 1 July — so if 1 July lands mid-cycle, the old rate still governs until the next period begins. A comparison benchmarked against last year's guide silently loosens; one benchmarked against a stale paste may never have been right.
The annual test is a floor, not a ceiling. Where a modern award's annualised wage arrangement applies, the annual reconciliation is not custom — it is required, along with records of start and finish times and unpaid breaks that the employee signs or acknowledges each pay period, and payment of any shortfall the reconciliation finds. Doing it once a year is the minimum the award asks for. The question is whether the minimum is still proportionate to the risk.
Most employers still answer that question annually, on typed-in data, against a hand-pasted floor. Some have already moved to quarterly or monthly, and a few have pushed the comparison into the pay cycle itself. We think the destination is further along than that.
What we built instead
Smartta OnTime now runs the BOOT test as a continuous control. Every week, for every configured employee, three inputs meet:
| Input | Manual spreadsheet | Continuous BOOT |
|---|---|---|
| Hours | Hand-typed from timesheets | Read directly from the clocking system of record — no transcription step to get wrong |
| Award floor | Pay guide pasted in, vintage unknown | Versioned Fair Work award pack, automatically period-correct — the annual increase applies itself from the first full pay period on or after 1 July |
| Employer rates | Typed into a cell | Entered once with mandatory provenance: who set the rate, when, from what source document |
| Cadence | One employee, once a year | Every configured employee, every week, with drift monitored |
| Output | A file on someone's laptop | A registered result per week, a drill-down to the day, and a printable per-employee report |
The result of each week is a verdict — and the verdicts are deliberately honest about what the system can and cannot see:
That middle verdict matters more than it looks. The single easiest way to destroy trust in a compliance tool is a false alarm — an automated "underpayment" that turns out to be a travel allowance the system couldn't see. Our engine treats its own blind spots as first-class:
A week without complete facts can produce a floor or a review. It never produces an accusation.
We mechanised the employer's judgment — we didn't replace it
There is no single canonical way to run this comparison; careful employers make defensible methodological choices, and reasonable advisers differ. So instead of imposing a house methodology, we did something more useful: we encoded our client's own method — every choice, including the generous ones — and proved the engine reproduces their manual workbook to the cent before letting it run forward. Their method, their judgment, on rails: correct hours, correct rates, every week, with every assumption written down where an auditor can read it.
Then we re-ran their original test period through the new engine with the complete facts supplied. All four weeks passed — the same verdicts they had reached by hand — now computed from the time system's records at the correct-year award rate, with the record discrepancies flagged for repair on both sides rather than silently absorbed. The same answer, standing on inputs everyone can defend. That, in miniature, is the entire point.
One control in a larger register
The BOOT test is the most visible piece of something broader. It only means anything if you first know which award applies — so it sits on a ratified, per-client instrument determination: a named human decides the covering award, records the reasoning, and pins the version of the Fair Work data it was decided against. Around it run structural checks (do your pay rules meet the award's shape — penalties, overtime tiers, casual loading), statutory checks (rest breaks, excessive hours, entitlement channels), and a governed exception register where every finding is opened by a monitor and closed only by a person, with a reason.
Together they form a workforce-governance posture built for the current era: continuously computed, honestly bounded, and evidenced end to end. When the question comes — from an auditor, a regulator, or your own board — the answer isn't a spreadsheet someone last touched in February. It's a register.
Going deeper: for the statutory Better Off Overall Test as the Commission applies it to enterprise agreements — and why the margin erodes across an agreement's life — see The BOOT Doesn't Stop at Approval.
The continuous BOOT test is part of the Payroll Compliance dashboard in Smartta OnTime — the same award-equivalent engine that supports s.193 BOOT monitoring for enterprise agreements (read that companion piece here). Client details in this article have been anonymised; the findings described are real, and their remediation is tracked in the client's governance register. This article is general information, not legal advice — award coverage, set-off drafting, annualised-wage clauses, IFAs and reconciliation methodology remain matters for qualified advice.
Sources & method. Statutory points are drawn from the Fair Work Ombudsman and the Fair Work Commission: the criminal underpayment offence commencing 1 January 2025 and the Voluntary Small Business Wage Compliance Code; the casual employee definition and employee choice pathway in force from August 2024; annual wage review increases applying from the first full pay period on or after 1 July; and the reconciliation and record-keeping obligations in modern award annualised wage arrangement clauses. The Better Off Overall Test proper is s.193 of the Fair Work Act 2009 and applies to enterprise agreement approval.