Microburbs Australian Property Data

Returns immediately if the API is up. No authentication required and not metered — safe to poll from a load balancer, uptime monitor, or status dashboard without affecting your usage. **Free.**

Takes no parameters.

Return the calling account and its current prepaid credit balance. **Free.**

Takes no parameters.

Microburbs ML AVM for a single GNAF — point estimate, 80% range, plus a confidence score (0–100) derived from the interval width. **Price: 10¢ per call.**

Bedroom count from the latest matching listing. **Price: 5¢ per call.**

Bathroom count from the latest matching listing. **Price: 5¢ per call.**

Parking-space count from the latest matching listing. **Price: 5¢ per call.**

Dwelling type (house, unit, townhouse, ...). **Price: 5¢ per call.**

Parcel land area in square metres from the cadastral feed. **Price: 3¢ per call.**

Resolve a GNAF to its full ABS Statistical Geography Standard hierarchy (Mesh Block → SA1 → SA2 → SA3 → SA4 → GCCSA → State) plus the parallel admin chain (LGA, suburb (SAL), postcode). Call this once to get every ABS key the rest of the data layers index on. **Price: 3¢ per call.**

All Sold + For Sale records, most-recent first. **Price: 30¢ per call.**

Most-recent Sold or For Sale record. **Price: 10¢ per call.**

All For Rent records, most-recent first. **Price: 30¢ per call.**

Most-recent For Rent record. **Price: 10¢ per call.**

Adjoining-parcel neighbours, enriched with address + property type + bed/bath/garage from the latest matching listing. **Price: 30¢ per call.**

Top comparable sales used to triangulate the subject property's value. **Price: 60¢ per call.**

Single best comparable from the CMA set (highest dist_score). **Price: 10¢ per call.**

Recent comparable sales in a small radius around the subject parcel. **Price: 4¢ per call.**

Primary zoning at the property's point — code (R2, B4, ...) when the council ships one, plus the human-readable name and the broader category. **Price: 3¢ per call.**

Development overlays intersecting the property's parcel. **Price: 5¢ per call.**

Schools within ~5km of the property, with rank + student counts. **Price: 5¢ per call.**

Schools within ~5km of the property, ranked by NAPLAN / socioeconomic score. **Price: 100¢ per call.**

Curated amenities (supermarkets, green spaces, restaurants/cafes) within `radius_m` of the property, bucketed and nearest-first. **Price: 10¢ per call.**

Nearest train / tram / bus / ferry stop to the property, per mode, with distance. Widen `radius_m` for the nearest station beyond 1km. **Price: 5¢ per call.**

Bushfire-prone-area overlays within 1 km. Read `on_property` — a returned row means "found nearby", not "this property is in a bushfire zone". **Price: 15¢ per call.**

Modelled flood overlays within 1 km. Read `on_property` — a returned row means "found nearby", not "this property floods". **Price: 10¢ per call.**

Recorded historical flood extents within 1 km. Read `on_property` — a returned row means "found nearby", not "this property flooded". **Price: 3¢ per call.**

Heritage-listing overlays within 1 km. Read `on_property` — a returned row means "found nearby", not "this property is heritage-listed". **Price: 3¢ per call.**

Public-housing concentration overlay near the parcel. **Price: 4¢ per call.**

Landslide-risk overlays within 1 km. Read `on_property` — a returned row means "found nearby", not "this property is at landslide risk". **Price: 3¢ per call.**

Coastal / soil erosion overlays within 1 km. Read `on_property` — a returned row means "found nearby", not "this property is eroding". **Price: 3¢ per call.**

Environmental-risk overlay (contamination, salinity, ...). **Price: 3¢ per call.**

High-voltage power-line easement near the parcel. **Price: 2¢ per call.**

Mobile-coverage black-spot overlay near the parcel. **Price: 2¢ per call.**

Every property-level risk overlay in one call. Flat 30¢ — may grow without increasing the bundle price as new overlays are added. With ``?geojson=true`` each risk that has a polygon overlay also carries its 1km-radius-clipped `geojson` FeatureCollection + `geojson_meta`. **Price: 48¢ per call.**

Our automated valuation for one property: the point estimate, the 80% range around it, and a 0-100 confidence score derived from how wide that range is. Today this bundles the single `/valuation/avm` call, so it costs the same; it exists so an agent can ask for "everything valuation" without tracking which valuation endpoints exist. Agent quoting behaviour, land values, negative gearing and the value series are separate calls. **Price: 10¢ per call.**

The physical facts about a dwelling in one call: bedrooms, bathrooms, parking spaces, dwelling type (house, unit, townhouse and so on) and land area in square metres. Priced as the sum of the five individual calls. Ask for a single field instead if that is all you need. **Price: 23¢ per call.**

Everything we hold on what this property has sold and rented for: the full sale history and the full rent history, each with its most recent record called out separately. This is the expensive one in the Property - History family, because it returns complete series rather than a latest figure. For "what did it last sell for", call `/history/sale/latest` instead. **Price: 80¢ per call.**

The evidence a valuer or agent would use to price this property: a CMA comparable set, the single closest comparable sale, and recent sales nearby. Use it to answer "what did similar homes around here actually sell for", and to show the working behind a valuation rather than a bare number. **Price: 74¢ per call.**

What the planning scheme allows on this block: its zoning code and every planning overlay that touches it (heritage, flood, bushfire, height and so on, as published by the council). Answers "can I build on it, and what is restricted". Development applications in the surrounding area are a suburb-level call. **Price: 8¢ per call.**

Schools for this address: the ones nearby with their distances, and how each ranks on NAPLAN alongside sector, level and attendance. The priciest bundle in the Property family, because it joins the school register, the ranking table and distance calculations. Call `/schools/nearby` alone if rankings do not matter. **Price: 105¢ per call.**

The suburb's upward ABS admin chain — LGA, SA2/3/4, GCCSA, state, postcode. Use ``/v1/suburbs/list`` to discover canonical SAL names (and filter by state / LGA / SA4 / postcode).

List Australian suburbs (SALs). Always returns `{suburb, state}` per row so SAL-name collisions across states are unambiguous. With no filter you get the first 1000 suburbs alphabetically (browser-renderable preview). Pass any of the filters to narrow: state / lga / sa4 / sa3 / postcode / q (name search). Filters combine (AND). Flat cost regardless of how many rows come back. **Name search (`q`) is typo-tolerant.** It tries an exact match, then substring, then a fuzzy near-miss, and returns `match` (`exact` / `contains` / `fuzzy`) and `score` on every row so you can see which happened. **Always tell the user which suburb you resolved to, and its state, before quoting numbers for it.** A fuzzy match is a suggestion, not a confirmation — `rokeby` is one letter from Kokeby in Western Australia and Rokeby exists in both Tasmania and Victoria. If more than one candidate is plausible, ask rather than pick. When `q` is combined with `state`, the state is a **preference, not a filter**: in-state candidates rank first, but a suburb of that name in another state is still returned rather than hidden, so a near-miss becomes "Seaview is in Victoria, not Tasmania" instead of "no data". Every other filter stays a strict AND. **Free.**

Latest median sale price for the suburb.

Full median sale-price observation history for the suburb. Well-covered suburbs extend back 20+ years; coverage and observation frequency vary. This is the raw historical series needed to calculate a stated 20-year return, not a precomputed "consistent growth" score or a forecast.

Latest median weekly rent for the suburb.

Full median weekly-rent observation history for the suburb, including 2024 and 2025 where the suburb has coverage. Older observations are generally month-end and recent observations can be weekly; this is not a uniformly quarterly series. Values are advertised median AUD per week.

5-year compound sale-price growth (decimal).

Past compound sale-price growth over 1 / 3 / 5 / 10 years, by property type (decimal). The 3y / 10y figures are the same ones the Suburb Finder ranks on.

Gross rental yield = median rent annualised / median sale price.

Median days on market for current listings.

Recent sale-transaction count for the suburb.

1-year sale-price growth forecast (low / most-likely / high).

Current for-sale listings in the suburb, with address, price, beds/baths, coordinates and an `off_market` flag per listing. `total` is the full number on the market (computed before `limit`), so a count is accurate even when the rows are capped. Raise `limit` for more listings; set `verbose=true` for the agent's full description text; set `off_market=true` for off-market listings only.

Rental listings live right now in the suburb, with address, weekly rent, beds/baths, coordinates and an `off_market` flag per listing. Same paging and filters as for-sale.

Population by age bracket and gender.

Household income breakdown by bracket.

Ancestry breakdown from the ABS census.

Country-of-birth breakdown from the ABS census.

Public-school catchment polygons covering the suburb.

Schools serving / near the suburb, with rank + attendance data.

Count of recent development applications in the suburb.

Recent development applications lodged in the suburb.

Share of the suburb AREA covered by bushfire-prone designation. Area, not properties — and the two differ a lot. Belmont North is 45.72% by area but 23.62% by property count, because designated land is not evenly built on. If you want "what share of homes here are affected", which is the figure our own reports headline, use `GET /v1/suburbs/{suburb_name}/risks/property-hazard-counts` — it returns a count and pct per hazard against the suburb's actual dwellings.

Share of the suburb AREA covered by modelled flood overlays. Area, not properties. For the share of DWELLINGS affected — the figure our own reports headline — use `GET /v1/suburbs/{suburb_name}/risks/property-hazard-counts`. To answer **where** the flood zone is, pass `geojson=true`. The response then includes the actual modelled flood polygons clipped to the suburb boundary, suitable for a street map, plus truncation/simplification metadata. Do not infer street location from `coverage_pct` alone.

Top-10 suburbs ranked closest by Microburbs' similarity model.

The eight headline market indicators in one call: median sale price and median rent (each with its full history), 5-year price growth, gross yield, days on market, and transaction volume. **This is a core set, not every market endpoint.** The `Suburb - Market` family has 29 endpoints; this bundles 8 of them, priced sum-of-parts. The path is `/market/all` for historical reasons — it is not "all". Supply, vacancy, growth drivers, affordability, mortgage stress, investor concentration, land values, repeat sales, bedroom stats and the per-mesh-block maps are separate calls, so you only pay for what you ask for.

Who lives in the suburb, from the census: population by age bracket and gender, and households by income bracket. Two calls bundled, priced as the sum. Tenure, employment, family make-up, schooling and the rest of the Suburb - Demographics family are separate.

Cultural make-up of the suburb from the census: ancestry breakdown and country of birth, each as a share of residents. Useful for understanding who a suburb's buyers and renters are. The map version, which splits these by mesh block, is a separate call.

Schools for a suburb: which public catchments cover it, and which schools sit nearby regardless of zone, with NAPLAN rank, sector, level and attendance for each. Catchments answer "where would my child be guaranteed a place"; nearby answers "what else is within reach".

What is being built in the suburb: the count of development applications on record, and the list itself with each application's address, type and status. Read it as supply coming down the pipeline, and as a signal of where a suburb is changing.

Natural-hazard exposure for the suburb: how much of its area sits inside a designated bushfire-prone area and inside a modelled flood overlay, each with a plain risk band and the number of distinct hazard polygons. Pass `geojson=true` to get the hazard shapes themselves, clipped to the suburb boundary, in the same call. Slope, heritage, power lines and the per-property hazard counts are separate calls.

Mesh-block codes inside a suburb. ``sal`` is required for now — nationwide / by-LGA / by-SA3 listings will land later when there's a story for the large payloads (a single LGA can hit a few thousand mesh blocks). **Price: 3¢ per call.**

The mesh block's upward ABS chain — SA1, SA2/3/4, GCCSA, state — plus the parallel admin chain (LGA, suburb (SAL), postcode). **Price: 3¢ per call.**

All Australian LGAs. Optional `?state=` filter (e.g. 'NSW'). Returns `{lga, state}` per row. Flat cost. **Price: 3¢ per call.**

LGA's parent chain — state + GCCSA. LGAs aren't strictly nested in SA4 so the upward chain stops there. **Price: 3¢ per call.**

All Australian SA4s. Optional `?state=` filter. Returns `{sa4, state}` per row. Flat cost. **Price: 3¢ per call.**

SA4's parent chain — GCCSA + state. **Price: 3¢ per call.**

Resolve a freeform address string to candidate G-NAF identifiers. Use this before calling any ``/v1/properties/{gnaf_id}/...`` route — those endpoints require a GNAF in the path. Returns up to 10 candidates, ranked by Meilisearch score. **Price: 1¢ per call.**

Autocomplete a partial suburb search string against the ABS SAL list. Use this before calling ``/v1/suburbs/{name}/...`` — those endpoints require a canonical SAL name in the path. **Free.**

Discover which statistical fields exist. Free (0c). Fuzzy, token-based matching covers table topics and clean column labels as well as terse field codes. Common natural-language plurals and Census synonyms are handled, so the verbatim `query=couples with children` discovers the exact 2021 ABS family-composition count fields rather than substituting the broader `census_family_households` metric. Use the returned field code(s) with the priced `/area-stats/SAL/{suburb}/values` route. ~17k fields available. **Free.**

List the regions that exist at a level — every suburb, LGA, postcode, etc. Free (0c). Use it to discover the exact region names to pass to /values. **Free.**

Statistical field values for one region. Pass `fields` — 1 to 5 field codes (from /fields) — to fetch. At SAL level, `mode=sa1` switches from the suburb-aggregate value to a per-SA1 breakdown — every SA1 that falls inside the suburb, each with its own values. **Price: 3¢ per call.**

State, SA3/SA4, postcode, population, dwellings and centroid for the suburb.

Nearest neighbouring suburbs with distance in km, closest first.

12-month sale-price growth by property type (house / unit).

Mortgage-stress score (0-100) for the suburb, with per-mesh-block breakdown.

Rental (investor) share per mesh block, with a suburb-wide rollup.

Estimated build-year profile — suburb aggregate + per-mesh-block medians.

Calibrated rental-vacancy rate for each mesh block in the suburb.

Suburb land-value medians ($/sqm, lot size, land share of price) + per-mesh-block $/sqm.

Median estimated annual strata levy for the suburb.

Microburbs affordability index for the suburb's postcode area.

House/unit price, sales volume, median bath/car/land, and stock mix per bedroom count.

Realised resale growth — summary + individual buy/sell pairs (max 500).

For-sale listings per 100 dwellings (90 days) — per mesh block + suburb rollup.

Median days on market across the suburb's greater region (GCCSA) — compare with /market/days-on-market.

Same-SA3 comparable-suburb candidates (canonical SAL names).

Median advertised weekly house rent for every microburb (mesh block) in the suburb, with its modal street name, plus the dearest and cheapest pockets. Values only — the matching polygons come from ``/shapes/mesh-blocks``.

Gross rental yield for every microburb (mesh block) in the suburb — weekly rent x 52 over median price — with both inputs returned so the number is checkable, plus the best- and worst-yielding pockets. Values only; polygons come from ``/shapes/mesh-blocks``.

All seven livability dimensions — affluence, lifestyle, hip, convenience, family, safety, tranquility — scored 0-100 for every microburb (mesh block) in the suburb, each with its human band. Averaging a dimension across the microburbs reproduces the suburb figure at ``/lifestyle/livability``. Values only; polygons come from ``/shapes/mesh-blocks``.

Properties sold in the suburb over the last 3 years, with beds/baths/land size, sale price and date, and coordinates. Paginated: `total` reports how many sales are available for the suburb (counted before `limit`/`offset`), so you can page through the full feed. Flat price per call regardless of page size.

Number of sales and median sold price in the suburb over the last 3 years.

Suburb and national total-crime medians (predicted incidents per 100,000 residents per year) with a vs-national verdict.

Per-crime-type rates (predicted incidents per 100,000 residents per year) for the suburb vs national medians.

Total crime rate for every mesh block in the suburb, keyed by ABS mesh-block code. Values are predicted incidents per 100,000 residents per year (see the payload's `unit` / `definition` keys).

Counts and shares of the suburb's properties flagged bushfire / flood / mine-subsidence / contamination / aircraft-noise / acid-sulfate.

Landslide / storm-tide coverage, EPA notices, landlord-insurance medians and crime benchmark, with per-microburb detail.

Counts of contamination points, heritage items and erosion-prone polygons inside the suburb.

Median / 90th-percentile property gradient and share of steep (>10°) properties.

Share of the suburb's properties inside a heritage polygon.

POI category breakdown, terrain scalars, nearest train/shop and beach/hospital flags.

What the absence of petrol stations, fast food and bottle shops near sold properties was worth.

Livability scores (0–100) per dimension, aggregated from mesh-block scores, with the SA4 benchmark.

Every named venue / amenity in the suburb as GeoJSON Point features, plus a per-category count. Each point carries `name`, a broad `category` (Restaurant / Cafe, Shopping, Health, Entertainment, Sport & Fitness, Education, …) and a finer `subtype` (e.g. `pub`, `bar`, `cafe`, `restaurant`, `gym`, `supermarket`, `school`, `pharmacy`). This is the endpoint for "what / how many <venue type> are in this suburb" and "name the pubs / cafes / gyms in X" — filter or count the features on `subtype`, and read real venue names from `name`. The top-level `categories` array gives the broad-category counts without walking the list. Unpaged by default. Big suburbs are big — Sydney is ~3.7 MB of features — so pass `limit`/`offset` (or narrow with `category`) when you don't need them all. `count` and `categories` always describe the whole suburb, so paging never hides the totals. Flat price per call regardless of page size.

Nearest beach and hospital with straight-line distances (null = none within ~50 km).

One plain-English sentence per livability dimension for the suburb.

Straight-line distance to the nearest city and to the nearest capital, in km. Both, because they answer different questions and are often nowhere near each other: Belmont North is 14 km from Newcastle and 108 km from Sydney. `nearest_city` is usually what a buyer means by "how far is town". The top-level `city` / `distance_km` fields are DEPRECATED. They still describe the CAPITAL, unchanged, so existing callers are unaffected — read `capital_city` or `nearest_city` instead.

Zoning polygons clipped to the suburb, with per-class and parent-group area shares.

Height-of-building and floor-space-ratio control polygons with band summaries.

Satellite-detected construction change per mesh block plus a suburb rollup.

New dwellings (last 24 months) split units vs houses per mesh block plus a rollup.

Dwelling-mix scalars — apartment share and average lot size.

Average annual residential council rate for the suburb's LGA.

Recent development applications (up to 100, newest first) with category counts.

Total suburb population is `current_population` (ABS Census 2021), alongside suburb census points (2011/2016/2021) and the LGA's yearly population series with a 10-year modelled projection tail. Use this route for questions such as "what is the total population of Clarkson?

ABN industry-mix pattern (Diverse / Professional-services-heavy / Mining-exposed / Low-commercial residential) plus the POI categories unusually common in the suburb, with named examples.

Projected first-preference shares per party, aggregated across the suburb's microburbs (booth-derived model — not an actual vote).

The dominant ancestry (and its share) in each of the suburb's microburbs, plus suburb-level ancestry shares. Pilot coverage.

Suburb cultural-integration composite (0-100) with national percentile and tier, plus the per-microburb scores behind it.

Explicit total owner-occupation rate plus owned-outright / mortgaged / renting / public-housing shares — a suburb rollup and the per-microburb downscaled Census shares. `owner_occupied_pct` is the requested headline; it equals owned outright plus mortgaged and is null if either is absent. `census_year` makes the 2021 reference period explicit.

How the suburb ranks inside its SA4 region and against its touching neighbours on renter share, welfare dependency, overseas-born parents and public housing.

Median weekly household income for every microburb in the suburb, plus the suburb median, national percentile and highest/lowest pockets.

2/4/8-year price forecasts for every street in the suburb, anchored to the real sold median. Unpaged by default. Large suburbs are large — Point Cook has 943 streets (~1.4 MB) — so pass `limit`/`offset` when you don't need the lot. `total` reports how many exist. Flat price per call regardless of page size.

Simplified SAL boundary polygon (GeoJSON Feature) plus centroid.

Neighbouring suburbs as a GeoJSON FeatureCollection — each with its boundary polygon and distance in km.

ABS 2021 mesh-block polygons covering the suburb (GeoJSON FeatureCollection, MB code per feature).

Hazard polygons for one layer, clipped to the suburb boundary (empty collection = hazard not present).

SA1 polygons in the suburb with their 2021 census public-housing share (only SA1s above 0%).

Precomputed noise colour regions clipped to the suburb, plus cell-centre heat points.

Registered easement geometries intersecting the suburb (GeoJSON FeatureCollection).

Busy-road line segments clipped to the suburb — the report's road-noise exposure proxy.

Council land valuations for the parcel — val_2021..val_2025 series with YoY change and $/sqm, plus lot area and zoning description. **Price: 15¢ per call.**

Modelled property value over time (monthly for the last year, annual before that). Houses only — units have no series. **Price: 30¢ per call.**

Negative-gearing stat at the finest available level: per-property cashflow scenarios (80% + 105% LVR), falling back to mesh-block then suburb investor-exposure. **Price: 20¢ per call.**

Quoting accuracy of the property's current listing agency — cohort, median guide-vs-sold deviation, underquote rate and its 5 most-recent tracked sales. Only while the property is listed for sale with a tracked agency. **Price: 50¢ per call.**

Every GNAF property within 200 m of the subject, with lat/lng and mesh block. **Price: 15¢ per call.**

ABS Census 2021 tenure shares for each mesh block around the subject. **Price: 15¢ per call.**

Properties around the subject flagged by the public-housing classifier. **Price: 15¢ per call.**

Cadastral parcel polygons (GeoJSON) for properties around the subject. **Price: 25¢ per call.**

Latest sold/rent dates and likely-owner-status signal per nearby property. **Price: 20¢ per call.**

Mesh-block polygons within 1 km of the property (GeoJSON), each with avg annual growth, investor counts, modal street and median sold price. **Price: 30¢ per call.**

The subject's bedroom band (bed count from listing history) and the suburb's 10-year price series for that band. **Price: 15¢ per call.**

Currently for-sale listings within `radius_km` of the property (newest first, max 30) with AVM and rent-yield comparison. **Price: 20¢ per call.**

Properties currently listed for sale in the suburb, split house/unit, with the suburb's rank among its peers.

How many months the current for-sale stock would take to clear at the suburb's current sales rate. Under 2 months is a seller's market.

Suburb-wide rental vacancy rate by property type, as a decimal fraction (0.00842 = 0.84%). Use `/market/vacancy-map` for the per-mesh-block surface.

Full gross-rental-yield history by property type, decimal fractions.

Full rental-vacancy history by property type, decimal fractions.

The last four weekly smart-median observations — house price, unit price and house rent — each with its week-on-week percentage change.

Whether high-voltage transmission lines cross the suburb, how many, and the highest voltage class (overhead lines only).

Matched-study capital-growth impact (%/yr) for bushfire, flood, contamination, mine subsidence, aircraft noise and surface acid sulfate, scaled by the share of properties affected.

Per-indicator capital-growth attribution for the suburb. Each row pairs the suburb's own value for an indicator with the capital-growth differential our matched studies attribute to it, in percentage points. Covers supply & demand pressure, market signals, environmental risks and demographic composition.

Welfare-dependency and public-housing shares with the suburb's underclass tier (High / Mid / Low) and the capital-growth impact associated with that tier. Regional and neighbour ranks for the same two metrics are served by `/demographics/cohorts`.

Labour-force-weighted unemployment rate for the suburb against the pinned national median, plus the per-microburb shares behind it.

Single-parent-family share for the suburb against the pinned national median, with the associated capital-growth impact.

Share of residents attending Catholic / Independent schools — suburb rollup plus the per-microburb shares behind it.

Dwellings and people per km², plus the land area and counts behind them.

Council capex / project-pipeline signals extracted from the LGA's meeting minutes — each with stage, dollar value, location and time horizon — plus a rollup of the total pipeline in dollars.

Median house price for every microburb (mesh block) in the suburb, with its modal street name, plus the dearest and cheapest pockets. Values only — the matching polygons come from ``/shapes/mesh-blocks``.

The property's own street: current median price, 2/4/8-year price forecasts, realised growth, sale/rental turnover, renter share and houses/units on the street — plus the suburb-wide medians to judge it against and the street's rank on 4-year forecast growth. 404s when the street has no forecast row rather than falling back to suburb-wide figures, which would answer a different question. **Price: 20¢ per call.**

Every field you can filter / sort on, with its label and unit. 0c. Houses and units are separate keys: the `unit`-prefixed ones (`unitMedianPrice`, `unitGrossYield`, `unitMedianRent`, …) are the apartment market, the unprefixed market keys are houses, and both can be filtered in the same query. Unit fields are NOT nationally populated — they only exist where a suburb has a unit market (roughly 13-54% of suburbs depending on the field), so a suburb with no unit market simply won't match a unit filter. **Free.**

Takes no parameters.

SA4 regions grouped by state — the values accepted by `states` / `regions` on search. 0c. **Free.**

Takes no parameters.

How many suburbs match the filter, without paying per row. 5c. Use this to size a result before searching: /search is capped at 25 rows per call, so `total` here tells you how many `offset` pages you need. **Price: 5¢ per call.**

Boolean-AND range filter over suburb metrics, single-column sort, paginated. Every row carries its `rank` — position in the ranked result (1 = top of the sort, out of `meta.total`). When sorting by a median or a growth, the actual value is also shown for the first 5 and last 5 rows of the page; for any other suburb's number, call the per-field endpoint (market/*, forecast/*). **Paging:** a page is capped at 25 rows. A bigger `limit` is clamped, not rejected — so always read `meta`: `total` is the full match count, `has_more` says whether rows remain, and `next_offset` is the `offset` to pass for the next page. A "top 50" needs two calls, each charged. Flat 30c. **Price: 30¢ per call.**