Where the numbers come from
The range is not an opinion about prices in the area. It is calculated from sale statistics three Australian governments publish, and this page records exactly which ones, for what period, and on what terms.
Three governments, three tables
NEW SOUTH WALES publishes the most, and it is the only one of the three that publishes a RANGE rather than a single figure. The Rent and Sales Report gives the first quartile, the median and the third quartile of recorded sale prices for each postcode, split into strata and non-strata, over the January - March 2026 quarter. 414 postcodes carry a usable figure, 228 of them with a strata split. Where New South Wales publishes a range, that range is what the widget shows: nothing is derived and nothing is widened.
VICTORIA publishes a MEDIAN by suburb, for houses, units and vacant land, over the Oct-Dec 2025 quarter, with the number of sales behind each figure. 779 suburbs are covered, 765 of them for houses and 442 for units.
SOUTH AUSTRALIA publishes a MEDIAN HOUSE price by suburb for metropolitan Adelaide only, over 2Q 2026: 373 suburbs across 19 councils. There is no unit table and nothing outside the metropolitan boundary.
A fourth dataset carries no prices at all. G-NAF, the national address file published by Geoscape Australia, is what places a typed address in its state, its suburb and its postcode. It is the address spine, and it is why an address in a state we cannot price is told which state it is in rather than treated as unknown.
The licences, in full
All three price sources and the address file are published under Creative Commons Attribution 4.0 International, which asks for the source to be named. The import script re-reads every licence on every run and refuses to write a new data file if one of them has changed, because a quiet move to a non-commercial licence would be invisible in the data and fatal in the product.
State of New South Wales (Department of Communities and Justice), Rent and Sales Report. Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution printed on the result: © State of New South Wales (Department of Communities and Justice)
State of Victoria (Department of Transport and Planning), Victorian Property Sales Report - Median House by Suburb Quarterly. Licence: Creative Commons Attribution 4.0 International (https://creativecommons.org/licenses/by/4.0/). Attribution printed on the result: © State of Victoria (Department of Transport and Planning)
Government of South Australia (Department for Housing and Urban Development), Metro median house sales. Licence: Creative Commons Attribution (http://creativecommons.org/licenses/by/4.0). Attribution printed on the result: © Government of South Australia (Department for Housing and Urban Development)
The New South Wales attribution is the department's own requested wording, and it asks that a link to its copyright notice be given with it: https://dcj.nsw.gov.au/statements/copyright-and-disclaimer.html
Why the answer is a range and not a number
Because a number would be made up. No open Australian dataset carries a floor area, a land size or a bedroom count alongside a price. Not one. So a two-bedroom cottage and a five-bedroom family home in the same suburb are described by the same published figure, and any adjustment between them would be a guess dressed as arithmetic.
What the range says instead is true and useful: this is what homes of this type sold for around here, in this quarter, according to the government that recorded the sales. On New South Wales it is the state's own middle half of those sales. On Victoria and South Australia the state publishes only the median, so the width around it is derived from two things that are measured rather than chosen: how far individual sale prices sit from a zone median in the New South Wales figures, and how far that suburb's own median has moved between published quarters. The wider of the two is used.
The widget also asks how many bedrooms, bathrooms and car spaces the home has, and how big it is. Those answers do NOT move the range, the result says so, and they are passed to the agency instead. An agency wants them; the arithmetic cannot use them.
Nothing here ever reaches high confidence, on any tier. That ceiling is structural rather than cautious: a range that does not know how big the house is cannot be a high-confidence estimate of that house.
Where there is no figure, and why
Sale statistics cover New South Wales, Victoria and metropolitan Adelaide: about 62% of Australians (Australian Bureau of Statistics, National, state and territory population, 31 December 2025, counting South Australia at its metropolitan share only).
QUEENSLAND, WESTERN AUSTRALIA, TASMANIA, THE NORTHERN TERRITORY AND THE ACT publish no open sale-price dataset that may be used commercially. Western Australia's Landgate sells its sales evidence under a paid contract; Queensland has twice publicly declined to publish one. New South Wales does publish an address-level sales file through the Valuer General, and it is licensed CC BY-NC-ND, which forbids both commercial use and derivative works, so it is not used here either. The widget refuses those addresses by name and shows nothing at all.
Within the three covered jurisdictions there are two more refusals worth knowing about. Where a state published a figure but withheld the number of sales behind it, which New South Wales does whenever thirty or fewer properties sold, the result says so and the confidence stays low. And where a suburb NAME is carried by more than one place in the same state, which is common in Victoria and South Australia, no figure is shown at all rather than the wrong suburb's.
Rents
The letting side is New South Wales only. The Rental Bond Board publishes weekly rents by postcode, dwelling type and bedroom count, from bonds actually lodged, over the April to June 2026 quarter, across 432 postcodes.
It is the one place on this market where an answer the owner gave changes the number: the rent tables do count bedrooms, and the result says so. Victoria and South Australia publish no equivalent that is read here, so a Melbourne or Adelaide letting enquiry is told that rather than given a national average.
What the estimate is not
It is a statistically derived estimate, not a valuation under the Australian Property Institute standards, and it must not be relied on as one. It is not an appraisal either: an appraisal is what a licensed agent gives after seeing the property, and it is the natural next step from here.
It describes the price level in a postcode or a suburb, not the condition, aspect, renovation or outlook of one particular home. Those are what the agent's visit is for.
A worked example of the difference: over the Oct-Dec 2025 quarter the Victorian figures span from the dearest suburb in the state to the cheapest, and the whole of that span sits inside the same published table. The suburb is doing the work, not the house.
How current this is
Each publisher sets its own cadence and all three are quarterly. The newest figures held here are: New South Wales sales, January - March 2026; New South Wales rents, April to June 2026; Victoria, Oct-Dec 2025; South Australia, 2Q 2026. The result card always names the period its own figure came from, so a seller reads which quarter they are looking at rather than assuming today.
The Victorian median moves between quarters, and how much it moves is measured rather than assumed: the width around a Victorian figure is drawn partly from the 5 published quarters that suburb's own median is known for.
Amounts on this market are written with the currency named, A$0 rather than a bare dollar sign, because a dollar sign on its own does not say which dollar.