Our rule
Public claims must not exceed shipped capability or available evidence
A repayment estimate is not a quote. A pathway screen is not a lender match. A lender match is not approval. An approval is not settlement. We label each state separately and correct content when evidence or regulation changes.
Editorial Standard
Each commercial pillar must answer a distinct Australian borrower decision. Supporting pages are retained only where the asset, seller, entity, evidence or transaction changes the practical answer. Pages that repeat the same advice with a different keyword are consolidated by permanent redirect.
Finance explanations are written or reviewed by a named subject-matter reviewer. Tax, legal, regulatory and government facts should link to primary sources such as the ATO, ASIC, RBA, PPSR or legislation. We distinguish:
- Fact: supported by a primary source or an identifiable product document.
- Professional guidance: practical interpretation based on experience, labelled as general information.
- Illustration: a worked example using disclosed assumptions, not an actual customer outcome.
- First-party result: an aggregated AFA dataset with period, sample size, filters and exclusions disclosed.
Indicative Rate Methodology
Rates published on AFA are planning ranges, not live lender quotes or offers of credit. A range must include an “as at” date, borrower assumptions and a statement that actual pricing depends on the applicant and transaction.
The review should consider available lender or aggregator information and the current funding environment. The RBA cash rate provides context but is not an asset-finance retail rate. We do not imply that a cash-rate change produces an identical movement in every lender's pricing.
Range limitations: advertised or observed rates can exclude fees, apply only to narrow prime profiles or use a comparison basis that differs from another product. Borrowers should compare the written rate, fees, term, balloon or residual, security, payout method and total amount payable.
Calculator Methodology
The repayment calculator uses standard amortisation from user-entered asset price, deposit, annual rate, term, payment frequency and balloon or residual. Establishment fees, ongoing fees, lender-specific day-count conventions, payout calculations and insurance are excluded unless explicitly shown.
Structure comparisons use disclosed planning assumptions and do not reproduce a lender contract. The calculator does not determine credit eligibility, servicing, approval probability, GST credits, depreciation, deductions, fringe benefits tax or instant asset write-off eligibility.
A larger balloon reduces scheduled repayments but leaves a larger final payment and will generally increase total interest. The borrower should stress-test a lower future asset value before relying on refinance or sale to clear the balloon.
Smart Match Labels
| Label | What it means | What it does not mean |
|---|
| Repayment estimate | Arithmetic from disclosed inputs | Rate quote or offer |
| Likely pathway | Explainable rules based on answers supplied | Live lender-policy match |
| Suitable lender match | Current, versioned lender policy evaluated against the application | Approval |
| Indicative offer | Lender-authorised pricing or terms | Formal approval |
| Approved | Lender has approved subject to the stated status and conditions | Settlement |
AFA will not describe the experience as AI-powered lender matching until current lender policies, reason codes, versioning and decision audit logs are live. Confidence must measure data completeness, not invented certainty.
First-Party Market Data
AFA will publish aggregated application or settlement insights only when the source period, metric definition and sample size are visible. Groups with fewer than 20 records are suppressed. Direct identifiers, free-text notes and contact details are never included in a public market dataset.
Applications, matches, approvals and settlements are different denominators and must not be mixed. Medians are preferred to averages for skewed values such as requested amount, time to decision and rate.
Public Market and Fleet Data
The AFA Research Data Lab uses official and explicitly attributed sources. Every derived dataset records its publisher, reference period, licence or copyright status, definition, calculation rule and known limitations. Registered fleet, RAV entries, retail sales, imports, investment, finance commitments, enquiries, approvals and settlements are separate measures and are never presented as interchangeable.
Buyer-type comparisons are published only when the source identifies buyer type and permits reuse. Open registration and RAV files do not identify private versus government buyers. Commonwealth fleet composition may be reported as its own population, but it must not be relabelled as government purchase share. Detailed VFACTS buyer-type data is not reproduced without FCAI permission or a suitable licence.
Private-market indicator modules may combine a rolling new-supply series with point-in-time registered fleet age, but the populations remain separately labelled. Manufacture year is not evidence that a vehicle was bought used, and a RAV entry is not evidence of a retail sale, private purchase or finance settlement.
When an official source does not publish the asset class buyers commonly use, AFA may create a versioned classification taxonomy. Those datasets publish the exact inclusion rule, ambiguous exclusions and coverage denominator. For example, the ute monitor matches exact light-commercial make/model pairs and does not relabel the complete light-commercial population as utes. Historical monthly monitors retain each source release so revisions and abnormal source values remain auditable.
Review and Correction Policy
- Core commercial and rate pages: reviewed when material policy changes and at least quarterly.
- Tax-year content: reviewed against enacted rules before the relevant period; announcements are labelled as proposals until enacted.
- Calculators: formula changes are tested against fixed examples before deployment.
- Material corrections: update the page date and remove or correct dependent structured data.
To report an error, use the contact page and include the page URL, disputed statement and supporting primary source.