Calculation methodology

Every calculator here is a pure, deterministic function separated from the interface. Here is how we build and check the maths.

Decimal-safe arithmetic

Money is never computed with raw floating-point numbers. We use decimal arithmetic and round explicitly at the point of display, with the rounding rule stated in each calculation’s breakdown, so cents add up the way an invoice or payslip would.

Versioned rules by financial year

Regulated figures — tax rates, the super guarantee, leave entitlements — are stored as effective-dated rule sets. Each carries its source, the date it takes effect, the date we last checked it, and a verification status. When a rule changes we publish a new version rather than editing history, so a result can always be traced to the rules that produced it.

Tested against official examples

Each engine ships with a test suite built from official worked examples and edge cases (zero and negative amounts, cents, thresholds, rounding). We also run property-based tests that assert relationships which must always hold — for example, that a GST-inclusive price always equals the base amount plus its GST.

Exact, conditional, or estimate

We label every result so you know how much to trust it. An exact result is unambiguous arithmetic. A conditional result depends on a rule or eligibility you told us. An estimate is a lender- or payroll-specific approximation the real figure may differ from. If an essential input is unknown, we ask or withhold the number rather than show a precise-looking guess.

See also our editorial policy. General information only — not financial, tax or legal advice.