How to research rental demand in an area

• Renter share and household composition from Census, with the Census night date. Tenure type “rented” in 2021 excludes rent-free dwellings and is not comparable to 2016 QuickStats.
• Dwelling structure and bedroom mix. Demand for two-bedroom apartments is not demand for four-bedroom houses.
• Age and education profile, as a hint about who lives there, not as a tenant credit file.
• Anchors: hospitals, campuses, large employers, defence or a CBD fringe. Confirm they still operate. See employment concentration.
• Letting times and ask reductions for the class you would buy, tracked over weeks. See achievable rent.
• New-lettings or bond series where that state publishes them, with the publisher’s exclusions. Victoria’s Rental Report uses bonds as a stock proxy and medians from new lettings.

Open the ABS QuickStats page for Brunswick (Vic.), geography SAL20361, Census night 2021. These figures are official and dated. They are not 2026 ERP, not a Districts median, and not a vacancy rate.

People: 24,896. Median age 34 (Victoria 38). Occupied private dwellings 11,040. Unoccupied private dwellings 1,594 (12.6 per cent). That unoccupied share includes holiday houses and dwellings empty for any reason. It is not rental vacancy.

Tenure of occupied dwellings: rented 5,305 (48.1 per cent) against Victoria 28.5 per cent. Separate houses 35.3 per cent, townhouses 29.3 per cent, flats or apartments 34.1 per cent (Victoria 12.1 per cent). Two-bedroom dwellings 43.1 per cent. Lone-person households 34.2 per cent and group households 14.1 per cent, both well above the Victorian shares. University or other higher education attendance 2,783 people.

What that file supports: in 2021 Brunswick had a deep rented stock, a much higher apartment and two-bedroom share than Victoria, and a younger adult and student-adjacent profile. What it does not support: that a particular three-bedroom house will let in 2026, or that rents will rise because renter share is high. Open the Districts Brunswick profile for the current advertised rent and sales counts, and keep those figures on their own period labels. Do not average them with 2021 Census rent ($441 a week).

A 48 per cent renter share does not tell you whether the two-bedroom apartment you are buying faces 4 listings or 40. Pair the Census mix with supply pressure for that class. Strong historical demand plus a large nearby completion pipeline is a market that may loosen. Modest demand with no pipeline can be steadier than a marketing story suggests. Direction over weeks beats a single snapshot.

• Treating 2021 renter share as 2026 demand.
• Mixing Census median rent with this week’s advertised median.
• Assuming a university headcount means demand for the house you are buying.
• Reading unoccupied dwellings as vacancy.
• Inferring demand from a low listing count in a small owner-occupier street.

It does not establish tenant quality, length of stay or that an anchor will still be hiring. It does not establish the rent for a specific dwelling. Census tables age. Bond series lag. Listing observations miss private lets. Where Districts has no household or tenure fields on a suburb profile, use the Census tables rather than inventing a Districts demand index.

Suburb profiles show population and, where a source publishes them, household or rental statistics with period and source. Property pages show comparable rental listings and the nearby pipeline. Industry of employment, tenure split and household composition are not always on the profile; the Census tables remain the official place for those. Districts does not compute a demand score or forecast rents.

Re-open the current Census or ERP product for the same geography code. Check the campus or hospital still exists and whether a large nearby project has completed since 2021. Track live listings for the class. Use that state’s bond or rental report only after reading its exclusions.