97% of Properties on the Same Street Move Together. City-wide? Just 71%.

Why does street-level data matter more than city averages?

“Sydney is up 5%.” That number covers millions of properties across thousands of streets. It describes a city. It tells you nothing about your street.

We measured this using annual median sale prices across every microburb in Australia over 15 years. Two houses on the same street moved in the same direction 97% of the time. Two houses in the same capital city agreed only 71.2% of the time. Nationally, it drops to 64%.

Most property data stops at the suburb. The 26-point gap between street and city is where that data fails you. Microburbs starts where that gap begins.

Detailed research: Market Cohesion

What we measure and how

How much are expert forecasts actually worth?

We collected national house price forecasts from major banks and economists for every year from 2009 to 2024. The consensus got the direction right 81% of the time.

A model that simply says “prices will rise” every year, with no analysis, would be right 75% of the time. Australian property prices go up in most years.

In 2023, PropTrack predicted national prices would fall 8% to 11%. Prices rose 5.5%. For a leveraged investor with an 80% loan, acting on that forecast would have cost 38.5% of equity.

We do not predict national direction. We focus on which streets and suburbs will outperform their region. Our top picks grew at 14.8% per year from 2009 to 2024, against a 7% market average.

Detailed research: Expert Forecast Reliability · Detailed research: Investor Performance

How does Microburbs compare on specifics?

Capabilities side by side, based on published documentation and our own testing.

Sources: RBA (Aug 2016), ACCC Digital Platform Services Inquiry (Mar 2024), PropTrack 2023 forecast (Broker Daily)

Detailed research: AVM Accuracy · Detailed research: LLM Forecasting Accuracy

How much does crime vary within a single suburb?

Every property portal shows one crime number per suburb. That treats every street the same. We used geocoded incident data from state police agencies to measure variation at the microburb level.

In Chatswood (NSW), the quietest microburb records 88 incidents per 100,000 residents. The busiest: 25,245 per 100,000. That is a 286x range within one suburb boundary. In Parramatta, the range is 63x.

The price gap between safe and risky microburbs in the same suburb averages 8.2%. On a $600,000 property, that is $49,000. We mapped this for 368,255 microburbs across all states.

Detailed research: Hyperlocal Crime Variation

What does public housing concentration do to growth?

No Australian government publishes which specific addresses are public housing. We built a classification model using tenure type, ownership records, and building characteristics to identify them at the property level across five states.

From 2020 to 2022, areas with zero public housing outgrew areas with 20%+ concentration by 14% to 27% depending on state. In Victoria, the gap was 26.8%. In NSW, each additional 10% of public housing concentration reduced two-year growth by roughly 4%.

Take Maroubra. The suburb average is 15% public housing. But the Coral Sea Park estate pushes some streets above 60%, while beachside blocks sit at zero. Two buyers 400 metres apart face completely different growth trajectories. Suburb averages hide that.

Detailed research: Public Housing and Property Prices

Can you get a price when the listing says “Contact Agent”?

At the time of our analysis, 27,282 of 93,510 for-sale listings across Australia (29%) hid the asking price. In New South Wales, the rate was 37.3%. In Tasmania, 7.4%.

Our AVM engine (6% median error rate, tested against 12 months of settled sales) produces a price estimate for 77% of those hidden listings. That is 21,081 properties. Portals that depend on agent-submitted data cannot fill this gap.

Detailed research: Price Transparency Study

How accurate is the agent's price guide?

We tracked 31,208 individual agents across 427,447 settled transactions from 2018 to 2024. 77.4% of agents had a median listing price below the final sale price. The typical gap was 3.5%.

On a $1 million home, that is $35,000. In Sydney's Inner West, agents underquoted by 7.9% on average. Only the Northern Territory showed a net overquoting pattern.

Detailed research: Agent Quoting Accuracy

What investors say after using the data

A real pocket-level example

One suburb had 28 pockets. All recent sales came from just 3 of them. The expensive ones.

Domain reported those sales as the suburb median. The whole area looked overpriced. But 25 of the 28 pockets told a different story.

We score all 28 pockets separately. You see which streets are genuinely performing and which are being lifted by a handful of sales on the other side of the suburb.

Where Microburbs falls short

We are a small team. That means real limits.

No commercial property. We cover residential only. If you are looking at office, retail, or industrial, we have nothing for you.

No rental yield tracking (yet). Our data covers sale prices, not rents. Rental yield estimates are on the roadmap but not in the current product.

Thin-market suburbs have wider error bars. In areas with fewer than 20 sales per year, our AVM accuracy drops and pocket-level scores become less reliable. We flag these areas, but they exist.

We do not predict national direction. We will not tell you whether “the market” will go up or down next year. Our model ranks streets and suburbs relative to each other. It does not forecast macro movements.

Coverage gaps in rural and regional areas. Our microburb mapping is densest in capital cities and major regional centres. Remote areas with sparse mesh blocks have less data and lower confidence scores.

Every claim on this page has a whitepaper behind it

Full data, methodology, and competitor analysis. No paywall. No email required.