One suburb can contain apartment precincts and large family homes. Mesh blocks, the smallest areas used by the Australian Bureau of Statistics, make those local differences easier to see.
Sold $765,800 · 15 Aug 2025
Street-level exterior of the Epping Park apartment complex, with stacked balconies and shared vehicle access. This is the building exterior, not a view inside Unit 371.
Source: Microburbs · Google exterior. Saved photo and dated listing facts. Camera date not supplied.
Sold $2,318,888 · 7 Dec 2025
A detached two-storey home with a double garage, front steps and a landscaped entrance.
Source: Microburbs · Google exterior. Saved photo and dated listing facts. Camera date not supplied.
Epping block A: retained listing records 16 Dec 2015–7 Jul 2026. Block B: 22 May 2016–25 Mar 2026. One dated housing record per observed address. These are selected clear contrasts, not a census of every home. Street images show exteriors and retain Google attribution.
Inspect the mapped properties and every source count →
42% of the suburbs in our test contained blocks with sharply different bedroom, bathroom or combined room profiles. This is a count of suburbs containing a contrast, not a claim about the same percentage of homes.
At least one pair of blocks met one of the large-difference definitions alongside.
5,779,724 property profiles across 171,195 blocks. At least 20 complete room profiles per block, 100 per suburb and two eligible blocks. Property-profile extract: 7 April 2026. Underlying observation dates are unavailable. ABS 2021 boundaries.
Bedroom mix: at least 70% of homes have 0–2 bedrooms in one block, while at least 70% have 4+ in another. 1,007 suburbs (21%).
Bathroom mix: at least 80% have one bathroom in one block, while at least 80% have two or more in another. 1,899 suburbs (39%).
Combined profile: at least 60% have 0–2 bedrooms and one bathroom in one block, while at least 60% have 4+ bedrooms and two or more bathrooms in another. 1,128 suburbs (23%).
The three groups overlap, so their percentages must not be added. This is a defined test of our available property profiles. Requiring at least 30 profiles per block gives 39% of qualifying suburbs, while 50 gives 27%. The minimum sample changes which smaller housing pockets can be included.
See the full denominators, definitions and sensitivity checks →
Yes. Werribee contains an established pocket dominated by three-bedroom homes and an estate where four-plus bedrooms and two-plus bathrooms dominate. “House in Werribee” still combines different stock.
Sold $555,000 · 1 May 2026
An established detached brick home with a tiled roof, front lawn and a driveway beside the house.
Source: Microburbs · Listing photograph. Saved photo and dated listing facts. Camera date not supplied.
Asking $450 per week · captured 28 Jun 2026
A single-storey home with a double garage, driveway and a small front lawn.
Source: Microburbs · Listing photograph. Saved photo and dated listing facts. Camera date not supplied.
Retained house records: A 3 Sep 2015–9 Apr 2026. B 8 Feb 2022–16 Jul 2026. Houses only. Apartments, townhouses, vacant land and unclassified records excluded. Each photo is an example, not a photograph of all homes in its block.
Inspect the housing mix, photographs and individual records →
Our modelled local household-income estimates indicate they can. 784 of 4,716 qualifying suburbs (16.6%) contain modelled local income profiles at least twice as high in one block as another.
Per household per week, a 3.3× difference. These are modelled local household-income medians. The published 2021 Census suburb median is $1,311 per week.
A suburb-wide figure conceals these estimated local differences. It cannot tell you which housing pocket an address belongs to.
Modelled A: 111 observed 2021 dwellings. B: 54. The photos elsewhere in this article do not establish the incomes of the people who live in those homes.
National income test: residential blocks with at least 20 observed Census dwellings and modelled income of at least $500 per week, in suburbs with at least five eligible blocks. The estimates have no supplied block-level accuracy bounds. They indicate possible local differences, not observed incomes or a description of every resident. The underlying modelled data extract is dated April 2026.
The 16.6% figure counts suburbs containing a twofold contrast. Only 0.35% of eligible blocks were themselves at least twice above, or below half, the published suburb income. Those are different questions and denominators.
See the modelled profiles, source definitions and exclusions →
Yes, in the measured house sample. Bedroom counts differed in 51% of suburb comparisons, against 39% for the nearest recorded house. Bathroom counts differed in 35% versus 26%. The same homes were used at every level.
Same 16,249 houses across 560 suburbs. Recorded listings: 3 January 2023–10 February 2026.
Shorter bars mean a higher share of homes have matching room counts. The comparison uses another recorded house in the suburb, another in the same mesh block, or the nearest recorded house within 100 metres.
The benefit of going local remains after restricting the comparison to houses. A mesh block preserves useful local housing detail, and a nearby individual house can be a closer match again. Neither removes the need to inspect the property.
16,249 houses across 560 suburbs and 1,705 mesh blocks, with unit-style addresses excluded. Same homes and same data at all three levels. Nearest sampled house more than 5 m and at most 100 m away, averaging 41.5 m. It is not necessarily the immediate next-door neighbour. At least 30 recorded houses per suburb and five other houses in the block. Retained listing dates: 3 January 2023–10 February 2026. ABS 2021.
This measures physical room similarity. It does not measure sale-price or rental-valuation accuracy, or prove that the statistical boundary itself caused the similarity.
See exact matching rates, distance tests and independent checks →
In Medina, south of Perth, the mapped examples show a clear change: five sampled houses built between 1953 and 1965 sit beside ten built between 2014 and 2016. All 15 have a reported construction year.
1953–1965 reported construction years. None of the block’s 51 currently registered addresses first entered the national address register in 2015 or later.
2014–2016 reported construction years. All 77 currently registered addresses first entered the register in 2015 or later.
The building years and address history both point to different development eras. Address registration alone does not prove when a home was built.
Construction-year source extract: 19 April 2026. Original observation dates not retained. National address-register history through 21 April 2026. Recorded houses are a sample, not every property in each block. The map shows locations. It does not establish building condition.
See all 15 addresses and their source construction-year statements →
Suburbs are useful for a broad conversation about city access or shopping centres. For an individual home, choose evidence that fits the actual property and its local housing stock.
Compare the home with competing rental stock of the same type, room count and condition. An apartment precinct and a large-family estate can have different tenants, layouts and competition.
A comparative market analysis should explain why each sale is relevant. Use the mesh block to understand the local stock, then select and adjust individual sales for date, land, condition and location.
Modelled income and household profiles can highlight areas worth investigating. Keep estimates separate from published Census counts and from what is known about a particular household.
Separate established housing, new estates and infill pockets. Track new addresses alongside planning and building evidence. Address registration dates alone do not prove construction dates.
Use the address and relevant mapped flood, road or environmental evidence. Keep crime data at its supported geography. Income estimates do not establish that an area or person is unsafe.
Use an official small-area code and boundary edition to join housing and local context consistently. Preserve the actual property details and each source date alongside them.
See application examples and limits →
Roads, property boundaries, land use, access and physical barriers all matter. A mesh block follows the real place. It is not a standard square.
Homes along a main road may share frontage and traffic exposure, while the houses behind face a quieter street. A small boundary can preserve that distinction for further investigation.
ABS guidance also seeks to keep housing complexes together, separate different land uses and group communities with shared access. It favours compact urban shapes where possible. Long strips are real examples, not a rule that every main road must produce one.
A road-shaped block is a useful clue, not a measured noise, safety or price effect. Read the ABS mesh-block design guidance.
They give different datasets a common, detailed geographic reference. You can preserve local variation first and combine it into larger official areas when needed.
The Australian Bureau of Statistics defines mesh blocks as the smallest areas in its Australian Statistical Geography Standard. Most contain 30 to 60 dwellings, although some have many more and others none.
SA means Statistical Area. Each level in this main structure combines whole areas from the level below. Other ABS structures use different groupings.
ABS approximations of suburbs, postal areas and council areas are also assembled from mesh blocks. Postcodes and suburb names remain useful labels, while a code and polygon make a data join explicit.
ABS standard and geographic hierarchy · Official 2026 allocation files
People recognise suburb and postcode names. They work well in conversations, spreadsheets and broad questions about city access or major shopping centres such as Westfield. Even then, the exact address determines the journey.
Discuss mesh-block data access with Microburbs
When a system compares thousands of properties, it does not need a memorable name for every housing pocket. It needs consistent boundaries, dated evidence and enough detail to avoid treating an entire suburb as one property market.
New development can turn one large statistical area into many smaller ones. Microburbs holds the 2026 mesh-block boundaries, so we can preserve this newer geography alongside the 2021 evidence.
In this selected Clyde North example, one 2021 mesh block became 43 successor blocks in 2026. That does not mean 43 new neighbourhoods appeared at once. It means the statistical boundaries now preserve more of the area’s detail.
This is a boundary and satellite comparison, not 2026 Census results. The first 2026 Census results are due in June 2027. ABS release plan.
Read the boundary correspondence and imagery dates →
The Microburbs API provides mesh-block identifiers, polygons and local property context, alongside property details and comparable-sales tools. That lets an application show the small area on a map and explain the home in its local setting.
For a buyer: Preview a suburb or property report. Full report detail requires a subscription. The public preview uses a clearly labelled sample location.
For a software team: keep the address identifier, mesh-block code, boundary edition and source dates together. Use the official geography to connect data without relying on suburb-name matching.
2026 boundaries are held by Microburbs. The public polygon API documentation currently names 2021 boundaries. Ask for the boundary edition you need when arranging access. This article does not announce a documented 2026 API endpoint.
Discuss mesh-block data access See the mesh-block API documentation
Preview a Microburbs suburb report · Property context API · Read the research note
Yes. Similar building forms can occur in different suburbs. That can help with stock and design comparisons, while local demand, amenities and land values still need separate attention.
A low brick house with a broad front garden, mature trees and a separate driveway along the side.
Sold $655,000 · 2 Dec 2025
Source: Microburbs · Saved realestate.com.au record. Camera date not supplied.
A single-storey exposed-brick house with a tiled roof, brick chimney, broad front windows and a front garden.
Sold $500,000 · 25 Sep 2025
Source: Microburbs · Saved realestate.com.au record. Camera date not supplied.
Selected three-bedroom, one-bathroom homes with brick walls, tiled roofs and front gardens. The individual dated sale prices differ. These are physical housing examples, not proof of interchangeable valuations or matching residents.
Inspect the cross-suburb comparison →
Chief Data Scientist and Founder
microburbs.com.au
