The ABS runs the Census every five years and it’s genuinely one of the best datasets in the country, but actually getting anything useful out of it is painful. Everything’s buried in DataPacks — hundreds of CSVs with column codes like G17C and Tot_P_P (you need a separate metadata workbook just to work out what you’re even looking at). I wanted to compare how a few suburbs had changed and ended up writing a parser, which is a ridiculous amount of effort just to answer “is my area getting older or younger”.
So I built CensusAtlas. It takes the 2011, 2016 and 2021 Censuses and turns the whole lot into maps, charts and plain-language profiles for every suburb, town, council area and statistical region in the country (around 90,000 areas all up). You can map median income or rent across Australia, see the most common ancestry or how people get to work in each neighbourhood, or just pull up your own street and watch how it’s shifted over the decade. No codes, no spreadsheets — pick a place and read.
A fair bit of the fiddly work is handled for you. Older years are re-aggregated onto current boundaries where the data allows, so the trends actually line up rather than jumping around every time the ABS redraws a suburb. And every number comes with context — rankings, distributions, nearby areas and how it’s changed over time — because a figure sitting on its own rarely tells you much. The whole thing is built to be fast and to stay cheap to run, so it can just sit there being useful.
It’s free at https://censusatlas.au — pick a topic, or open the map and zoom to somewhere you know. All the data comes straight from the ABS Census (used under CC BY 4.0); it’s an independent project and isn’t affiliated with them. There’s still a pile of Census data I haven’t surfaced yet, so if there’s a statistic you wish it showed, there’s a contact link on the site — let me know.