Area explains the growth. This explains the area.
HtAG already ranks 7,000+ suburbs on measures you weight yourself. StreetLens takes the same idea below the suburb boundary: every suburb divided into cells of identical size and shape, each with its own price, rent, yield, days on market, affordability, hold period, land, age, tenure and hazard reading — and the same significance dials, so the map ranks by what matters to you rather than by what matters to us.
| Measured | Result |
|---|---|
| Price gap between the cheaper and dearer parts of one suburb | $297,024 |
| The same gap, middling areas only | $139,273 |
| Suburbs with a 30%+ internal price gap | 67% |
| Suburbs where flood exposure is mixed | 43.9% |
| Suburbs where bushfire exposure is mixed | 47.0% |
| How much yield varies inside a suburb | 0.34pp |
Most property scores are someone else’s opinion, and they don’t tell you whose
Australian property analytics has settled on a convention: gather a set of measures, combine them into a single number, and keep the recipe. One well-known platform describes its own method, on its own homepage, as a “secret weighting”. That is the norm, not an outlier — and it is a defensible way to build a product, so long as you and the vendor happen to want the same thing.
| The usual arrangement | What StreetLens does instead |
|---|---|
| One composite score, one recipe, applied to every buyer | You set the significance of each measure, so a yield buyer and a growth buyer get different maps of the same ground |
| The weighting is fixed, undisclosed, or both | The weighting is visible, adjustable and yours — and it travels with the link you share |
| Weighting decisions, where they exist, stop at the suburb | The same dials apply below the suburb boundary, where the variation actually is |
| Every measure treated as equally important — mathematically the same as treating none of them as important | Turn a measure to zero and it leaves the ranking entirely |
There is no single correct weighting across a yield buyer and a growth buyer, so we don’t ship one. The honest version of a composite score is one where the person who carries the risk sets the priorities.
Same cells. Different priorities. Different winners.
Below is one set of cells rendered twice, on a loop. Nothing about the ground changes — only which measures were told to matter. The cells that lead under a growth weighting are not the cells that lead under an income weighting, and that disagreement is the entire point.
The map re-ranks as the dials move. Illustrative — a demonstration of the mechanism, not live data.
The same map answers a different question depending on who is asking
Below the suburb boundary the useful question stops being generic. An investor is trying to settle an argument, a buyers agent is trying to apply a brief, and a broker is looking at security risk. Those are three different uses of the same fourteen measures.
Suburb research gets you to a shortlist and then stops. StreetLens keeps going: inside one suburb you get a neighbourhood-by-neighbourhood surface, the live for-sale listings plotted on it, and filters for beds, baths and price so you are looking at the actual stock in the actual part you would buy in. Ask any forum whether yield, vacancy or supply matters most and you get four confident answers — so set the weighting yourself, start from a preset like High yield or Low risk, and the disagreement becomes a stated assumption you can inspect and change.
What it replaces: the three or four free map sites people currently cross-reference by hand for tenure, socio-economic and hazard layers, plus the spreadsheet that stitches them together.
You already weight a brief to rank suburbs. This does it a layer down: the same fourteen measures resolved to neighbourhood cells inside the suburb, the client’s live listings scored against that surface, and a split view to hold two markets side by side. The harder question stops being which suburb and becomes which measures deserve weight for this brief — and what to do when two disagree inside one suburb, where the affordable part is also the exposed part.
What it gives you: the client’s brief expressed as a visible set of priorities, exportable with your own legend, and a link that opens on the exact view you were looking at.
A suburb average can sit comfortably over a security that isn’t in the comfortable part, and the suburb report will never show you that. At neighbourhood resolution you can see where a hazard overlay cuts through a suburb rather than covering it — insurer appetite and valuation risk change on one side of a road and not the other — and check the specific listing your client is financing against the cells around it.
Stay in your lane: this produces evidence to hand a client, not advice to own. It is a property-information tool and nothing on this page should be read as a statement about credit obligations.
None of them is short of data. All three are short of a defensible way to say which data mattered and why — and to show that reasoning to somebody else afterwards.
See plans and pricingHow much variation is actually down there, measured rather than asserted
Computed from HtAG’s own cell-level tables — every mapped cell in the country, houses, all bedroom and bathroom counts, cells carrying at least two backing properties. These are research samples with the sample size stated, not platform claims.
| What was measured | Result | Sample |
|---|---|---|
| Dearer areas vs cheaper areas of the same suburb, as a ratio (90th vs 10th percentile cell) | 1.40× — a 40% gap | 8,571 suburbs with 5+ mapped cells |
| The same gap, in dollars | $267,571 | 8,571 suburbs |
| The same gap, restricted to genuinely suburban blocks of 200–1,200 m² — the robustness check | 1.417× · $297,024 | 881 suburbs with 8+ cells |
| The conservative version: middling areas only, discarding the top and bottom quarter | $139,273 — an 18.7% gap | 881 suburbs |
| Suburbs where flood exposure is mixed — some parts exposed, some not | 43.9% | 13,519 suburbs |
| Suburbs where bushfire exposure is mixed | 47.0% | 13,519 suburbs |
| How much gross yield varies inside a suburb | 0.34pp — almost flat | 881 suburbs |
The uncontrolled figure (1.402×) and the lot-size-controlled figure (1.417×) are almost identical — which is the point of running both. The gap is not an artefact of large landholdings classified as houses.
One caveat we would rather state than have found. Ranking suburbs by their most extreme cells produces implausible results — englobo land and development sites priced as houses. The extremes are not published here and should not be quoted. The percentile measures above are the defensible ones.
“Go fine enough and the numbers stop meaning anything”
This is the right objection and it has been made in public, for years, by people who know what they are talking about. Below a certain sales volume a growth rate is noise wearing the costume of a number. Here is exactly what we do about it, rather than a footnote saying we thought about it.
One unusual sale cannot move a cell. Every figure is a median of the properties inside it, which is why a single trophy home in an ordinary street doesn’t repaint the area around it.
A cell needs a minimum number of backing properties before it renders a figure of its own. Below that it stays blank rather than inventing a number from one sale.
Where a cell is too thin to stand alone it shows the figure from the larger area containing it, and says so. It never fabricates a local number to fill a gap.
If a measure is drawn from a thin set, the map says so instead of looking equally confident everywhere. Different measures also resolve at different scales — social and hazard readings sit finer than market ones.
The same discipline that keeps 7,000+ suburbs in the platform and leaves the rest out applies one level down. Australia has more than 15,000 suburbs; fewer than half transact often enough to be analysed honestly. A confident figure drawn from three sales a year is worse than no figure at all.
Equal-area cells are what make a weighted score mean anything
Every other sub-suburb unit in this market is irregular — bounded by roads, drawn from clustering, or inherited from an administrative boundary that varies in size from one end of a city to the other. That is not a detail. The unit you choose changes the answer you get.
| The property of the grid | What it makes possible |
|---|---|
| Every cell is the same size and shape, anywhere in the country | An area in one city is a comparable area to one in another. Rankings mean something across markets, not only within one. |
| Cells nest cleanly inside larger cells | Zooming changes resolution without redrawing boundaries, so a figure at one zoom stays consistent with the figure above it. |
| Every measure is indexed to the identical cell | Weighting is mathematically coherent — you are combining measures that describe exactly the same ground, not overlapping approximations of it. |
| Boundaries are not drawn from the data | A cell can’t be shaped, deliberately or accidentally, to flatter the number inside it. |
We are not claiming to be first below the suburb — others publish there, and some at a finer unit than ours in dense inner-city areas. What the uniform grid buys is comparability and the ability to weight coherently, and that is the claim we will defend.
Price is free to anyone. The other thirteen measures need a plan.
Stated plainly, because a map that hides its gate wastes your time. Open any suburb dashboard without logging in and the price map draws for you — the real thing, not a sample. Every other measure is visible in the switcher and badged with the plan that opens it.
- All fourteen measures on every mapped cell
- My Score — set the significance of each measure yourself
- Live for-sale listings inside the cell you’re looking at
- Draw your own shape and get the median inside it
- Export the map with your legend for a client report
- Share a link that opens on your exact view and weighting
- All 7,000+ Australian suburbs and 537 LGAs
- Full price measure on every mapped cell
- Zoom, pan and click any cell for its price
- Every other measure listed with its plan shown
- No trial, no card, no login required
Pick a state, then a suburb, then set what matters
StreetLens sits on every suburb dashboard. Nothing to install and nothing to configure — choose a market and the map is already there.
The questions we actually get asked
Decide what matters before you commit the capital
Pick a state, open a suburb you know well, and see whether it looks as uniform inside as its median implies. The price measure needs no account.
Start with a suburbRent, yield, days on market, affordability, hold period, land, age, listings and My Score on Investor. Hazard, socio-economic and tenure on Professional.
See plans and pricingReuse any figure on this page, with attribution
HtAG Analytics (2026). StreetLens: weighting property measures below the suburb boundary. Available at https://www.htag.com.au/streetlens/