Summary

  • The ACCC’s proposed retail record-keeping rule would link individual broadband, fixed-voice and mobile services to plan, location, access and billing records.
  • Its financial fields measure charges issued on bills, not verified payments. That is a sharper view of what providers charged than an advertised plan list, but it cannot by itself show cash outlay, hardship or service quality.

A price list loses the customer

An advertised plan is a promise about a product. A customer’s bill is the place where the promise meets discounts, usage, add-ons and fees. Neither record alone explains the whole market. A public plan page may show a monthly price but miss a legacy offer still used by thousands of customers; a bill may show a total without explaining which plan, network or bundle produced it.

The Australian Competition and Consumer Commission (ACCC) is proposing a way to join those views. Its Retail Telecommunications Record-Keeping Rule would require specified carriage-service providers to retain and report records for retail broadband, fixed voice and mobile services, alongside separate plan records. A plan ID would connect a service to the terms and features of the offer. The ACCC could then compare the plan’s stated price and speed with the charges and access product attached to the active service.

That is a material change from counting services within market categories already defined by existing rules. The consultation paper says 11 retailers currently report counts of retail services. The proposal would move toward service-level records, allowing the regulator to aggregate them into geographic and product markets instead of being limited to the categories already set in a rule. The ACCC project page still labels the work “Ongoing”; the proposal is not yet a binding instrument.

The bill is the observation, not the payment

For broadband and mobile services, the draft asks for the last bill and the total of bills issued during a reference period. It would include discounts and additional charges, with separate fields for usage, add-ons, other fees and discounts. A separate plan table would preserve the monthly plan price before discounts and extras. The point is to see more than a list of advertised tariffs: what was billed for a service, under which offer, and with which extras.

But the accounting boundary matters. The proposal operationalizes a price as an amount billed. It does not require a field confirming that the customer paid the bill, whether the payment was late, whether a credit later offset it, or whether the household carried an arrear. The ACCC’s explanatory paper distinguishes the price charged in a bill from the amount paid. Section 151CM of the Competition and Consumer Act describes the regulator’s annual reporting duty in terms of charges “paid” by consumers; the draft’s bill fields are a proposed way to observe charges, not a verified settlement ledger.

That distinction does not make the dataset useless. Billed amounts can reveal how advertised prices are modified by discounts, top-ups, excess usage and fees. But they cannot be silently relabelled as out-of-pocket spending. Nor can they alone establish whether a plan is affordable for a particular household, whether a higher bill reflects greater use, or whether a service delivered its advertised quality. Those questions require additional evidence and a clear account of how the ACCC turns billing records into published comparisons.

A map of offers in use

The proposed service record would carry a scrambled customer ID, provider brand, plan ID, service start date, location and access-network information. For broadband, that includes the carrier and access technology; the record would also identify the previous provider where known. The plan table could include both the wholesale speed product purchased and the speed advertised to customers. Linking the two gives the ACCC a way to examine retail offers in the context of the underlying access product rather than treating a speed label as a complete description of service.

Geography is useful, but it is not a map of actual usage. The proposed mobile location is the SA1 containing the billing address. That can support regional comparisons, but it does not show where a subscriber made calls or used data. For fixed broadband, the SA1 identifies the service location. The difference should remain visible in any analysis that compares mobile and fixed markets.

The rules would cover specified providers, not every seller in Australia. The proposed list includes a 50,000-services-in-operation criterion for adding providers, while preserving some smaller-market providers such as Sky Muster retailers. The ACCC is also asking whether roaming satellite broadband, including Starlink Roam, belongs with fixed broadband or mobile reporting. These boundaries determine which services appear in the dataset, and therefore which comparisons it can support.

The records are also periodic. The draft proposes reference periods in March and September, with reports due six weeks after each reference date. Services are reported if supplied on the reference date; a previous provider is reported where known, but the ACCC does not propose requiring providers to reconstruct unknown histories. Plans with no active customers would not be reported. The result would be a recurring snapshot with a switching signal, not a complete event stream of every connection, cancellation or transaction.

Privacy and operating cost sit inside the design

A scrambled customer ID is meant to let the ACCC connect records across reporting periods without receiving a customer’s direct identifier. The draft says the value must remain consistent for that customer across reports while being unusable to identify the person or derive the underlying customer ID. That stability is what makes longitudinal switching analysis possible; it is also why the method and access controls matter. “Scrambled” should not be used as shorthand for anonymous or risk-free.

The ACCC says that because the proposal no longer requires personally identifying information to be reported, it did not conduct a full privacy impact assessment. It also says the records remain protected as commercially sensitive information under section 155AAA of the Act. The two points belong together: direct identifiers may be excluded, while linked billing, location and switching histories still require careful handling. The draft proposes a two-year retention period.

There is a parallel cost question. The current consultation paper’s summary of the earlier 2025 review records industry concerns about the work needed to create per-service records and link billing and plan systems. It says providers previously estimated six to 18 months to build new processes. Those are views from the 2025 review, not submissions to the consultation that closed on 7 October 2026. The proposed first retail reference period is September 2027, after a planned final instrument in Q4 2026. The schedule gives providers time, but the final fields, thresholds and reporting format will determine whether that time is adequate.

What the proposal can establish

The statutory authority is clear: section 151BU allows the ACCC to make a written instrument requiring specified providers to keep and retain records, and to prepare reports from them. That power turns data definitions into a control surface. The choice of providers, plan IDs, location, billing fields and reporting dates shapes which parts of retail competition become visible to the regulator.

The 2026 consultation sought views on those details, including burden and record formats. It closed on 7 October. As of 13 October, the ACCC’s project page still says “Ongoing,” and the checked official pages show no final instrument. Consumer groups, providers and government can contribute evidence; their participation does not itself decide the rule. The final instrument and the ACCC’s reasons will carry that authority.

If adopted, the proposal could make it easier to compare a bill with the offer and access product behind it, across regions and over time. It would not transform a bill into proof of payment or a price comparison into a verdict on household welfare. The ledger’s value will depend on preserving those boundaries in both the data and the public claims drawn from it.

Sources