What’s on the site
| Section | What it does |
|---|---|
| Suburb guides | A page for each of 243 Greater Brisbane suburbs: growth, median rent, who lives there, commute and what to watch out for, compared with the Queensland average |
| Suburb matcher | Four questions (direction, budget, household, must-haves) that shortlist suburbs across 8 regions, with the trade-offs for each |
| Schools and catchments | Every state, Catholic and independent school with catchments and background data, deliberately never ranked, plus a catchment finder |
| Rankings and comparisons | Fastest-growing, cheapest rent, youngest and more, plus side-by-side suburb comparisons |
| Cost of living | What's getting more expensive in Brisbane, and Brisbane against Sydney, Melbourne, Perth, the Gold Coast and the Sunshine Coast |
| Study and lifestyle | Universities and the suburbs near each campus, TAFE, parks and trails, commute times and a history of Brisbane |
The idea
People moving to Brisbane ask the same questions: is this suburb growing, what does rent cost, which school am I zoned for, how long is the commute? The answers exist, but they're spread across spreadsheets from the ABS, state agencies and council. I wanted one page per suburb that answers them the way a local would.
How it’s built
A scheduled job pulls data from the Australian Bureau of Statistics, the Residential Tenancies Authority, Queensland Police, Brisbane City Council and OpenStreetMap, then maps it to 2021 suburb boundaries. Each suburb is scored against the state average, and AI turns those numbers into plain-English pros and cons. Next.js renders every page as plain HTML so it loads fast and search engines can read it.
Stack Next.js · ABS Data API · Postgres · LLM API · Recharts · Netlify
Skills it shows
- AI development: a pipeline where AI writes 243 suburb summaries from structured data, not from guesswork.
- Analytics: joining five public data sources, benchmarking each suburb, and publishing the known gaps (crime by police division, flood risk sampled on a grid) openly.
- Product thinking: a matcher and catchment finder built around the questions people actually ask.
What I learned
- Being upfront about data limits builds trust faster than polishing over them.
- AI writes good summaries only when the numbers underneath are clean.
- People share a sentence, not a chart. Plain English does the work.
Behind the scenes
Why I built it
The goal is simple: answer the questions people actually ask when weighing up a suburb, from the source data rather than from sales material. Is it growing? What does it cost to rent? Who would the neighbours be? Is it flood prone? How long is the trip into the city?
Most suburb information online comes from real estate portals, where the job of the page is to sell a listing. The raw facts do exist, published by the ABS, the Queensland Government and Brisbane City Council, but they're scattered across dozens of portals, spreadsheets and APIs, and written for statisticians. I wanted one place that pulls them together and explains them like a friend who happens to read census tables.
The main audience is people relocating, especially those moving to Australia from overseas who don't know Calamvale from Chermside. Renters, buyers, families comparing school catchments, and students choosing where to live near campus are close behind.
The site has no property listings and never ranks individual schools. It's just the public data, explained.
The methodology
A site like this is only worth anything if people can trust the numbers, so I set a few rules early and built the whole thing around them.
- Only published data, never invented data. Every figure in a data section comes from an open dataset and is credited on the page where it appears. If a suburb is missing a dataset, the page says so or leaves the section out. Nothing is modelled or filled with a placeholder.
- Never shown in isolation. Every suburb is benchmarked against the Queensland average, so "median rent $690 a week" always comes with the context of whether that's high or low.
- Plain English first. Terms like SEIFA, ICSEA and median household income are explained inline, where the reader meets them, not in a glossary.
- Every chart has a table. Each chart is paired with a one sentence summary and a collapsible data table, so the numbers are readable without the picture and accessible to screen readers.
- Dates are always visible. Census data is five yearly, population is annual and rents are quarterly. Each page states exactly what date its figures are current to.
- Approximations are named, not hidden. Where a dataset doesn't line up neatly with suburb boundaries, the page explains how it was matched and what that means.
- Logic lives in code. Comparisons, rankings and "which suburb wins" calculations are deterministic code, not AI generated text, so the same inputs always produce the same answer.
All of this is documented publicly on the site's About and methodology page.
Where the data comes from
The site draws on more than 15 public datasets from federal, state and local government, plus OpenStreetMap. Almost all are published under a Creative Commons licence that allows reuse with attribution.
| Source | Publisher | What it powers | Refresh |
|---|---|---|---|
| Regional Population | Australian Bureau of Statistics | Population, growth, migration | Annual |
| Census 2021 | Australian Bureau of Statistics | Income, rent, mortgages, housing, work, education, households, birthplace, language | Five yearly |
| SEIFA 2021 | Australian Bureau of Statistics | Socio economic advantage and disadvantage | Five yearly |
| Consumer Price Index | Australian Bureau of Statistics | Cost of living trends by category and city | Quarterly |
| Median weekly rents | Residential Tenancies Authority (Qld) | Rent by dwelling type and bedrooms | Quarterly |
| Reported offences | Queensland Police Service | Crime rates by category against the state rate | Monthly |
| Schools directory and catchments | Queensland Government | School details and state school catchments | Yearly |
| School Profile (ICSEA) | ACARA | School socio educational context | Yearly |
| Flood Awareness, parks and trails | Brisbane City Council | Historic flood extents, flood likelihood, parks and trails | As updated |
| SEQ timetable (GTFS) | TransLink | Public transport time and frequency to the CBD | A few times a year |
| OpenStreetMap | OpenStreetMap contributors | Amenity counts and map tiles | As updated |
| National VET Register and TEQSA register | Australian and Queensland Governments | Universities, TAFE and training providers | Yearly |
| ASGS SA2 boundaries | Australian Bureau of Statistics | The suburb outline on every map | Five yearly |
Automated jobs pull each source on its own schedule, and the site is rebuilt as new releases land. Property sale prices, NBN availability and development applications aren't covered yet, and the site says so openly rather than leaving them out silently.
The hard problems
The interesting work wasn't building pages. It was making datasets that were never designed to fit together line up honestly.
Suburbs aren't one thing. The ABS publishes by statistical area (SA2), the rental board by postcode locality, police by police division, and council flood maps as freeform shapes. None of them share a boundary. Each source needed its own matching method, and each method is disclosed on the page.
Crime data at the right level. Police publish offences by division, a catchment that spans several suburbs. Rather than divide it up and present a made up suburb figure, I kept suburb codes off the crime table entirely and only join through a hand curated division to suburb mapping. The database structure itself makes it impossible to fabricate a suburb level number.
Flood risk without a per property lookup. Council flood likelihood shapes aren't clipped to any boundary, so I lay a roughly 40 metre grid over each suburb and sample it. It's an approximation, and the page says exactly that.
Rents that keep disappearing. The rental board replaces its quarterly file in place with no archive, so the pipeline saves a copy at every ingest to build a history over time. Where one postcode locality spans several suburbs, the medians are combined and weighted by the number of bonds lodged.
Two censuses, two maps. Comparing 2016 to 2021 means mapping old figures onto new boundaries via the ABS Time Series Profile, which is approximate wherever a boundary moved. The 2026 Census was held in August, so a full refresh plan is ready for when that data lands in 2027.
Thin data shouldn't look solid. OpenStreetMap undercounts cafés and GPs in less mapped areas. Below a threshold, the page says a category isn't well mapped instead of printing a misleadingly small number.
Technical SEO and organic traffic
There's no ad budget behind this site, so organic search is the growth plan. My approach is to build pages that answer the exact questions people type, make them easy for search engines to crawl and trust, then keep improving based on what Search Console shows.
Match pages to real searches. People don't search "SA2 demographic profile". They search "is Sunnybank Hills a good suburb", "fastest growing suburbs in Brisbane" or "Sydney vs Brisbane cost of living". Every suburb, ranking, school, catchment and city comparison gets its own page, targeting a long tail of specific questions instead of a few competitive head terms.
Organise content into clusters. Pages are grouped around four themes: suburb living, schools and catchments, cost of living and relocation, and transport and commute. Strong internal linking between suburbs, nearby suburbs, schools and rankings helps search engines understand the topic depth and lets readers keep exploring.
Get the technical foundations right. One canonical domain with matching redirects and canonical tags, clean XML sitemaps, descriptive titles and meta descriptions, structured data, fast pages that work well on mobile, and a unique social share image for every suburb.
Get indexed quickly, everywhere. Beyond Google Search Console, the site is verified in Bing Webmaster Tools and uses IndexNow to notify search engines the moment pages change. Bing's index also feeds a number of AI search tools, which matters as more people ask an assistant instead of a search box.
Earn trust. Relocation and money decisions are high stakes, so trust signals carry weight. The site has a named author, a public methodology page, sources credited on every page, and clear data dates.
Build things worth linking to. Data pages rarely earn backlinks on their own. Guides, a Brisbane history section, the suburb matcher and upcoming calculators are designed to be useful enough that blogs, forums and relocation communities link to them.
What's next
- Calculators: a rent bond calculator, then stamp duty, then rent versus buy, built in that order so each reuses the last one's logic. Every Queensland rate and threshold is pulled from official sources with a recorded verification date, never hardcoded.
- Property prices: suburb level sale price estimates, the biggest gap the site openly lists today.
- More hazard and planning data: bushfire, landslide and coastal overlays, plus council zoning, heritage and future infrastructure.
- More cities: Live in Sydney is live, and the same foundation can extend further.