How to compare suburbs for property investment

Two suburbs can look equally good on paper and still be very different places to own a property. A suburb score is a number from 0 to 100 that ranks a suburb against the others in its own state, so you can see where it sits among its peers.

It does not forecast a return, value a property or say which suburb to choose. A fair comparison puts two suburbs side by side for the same home type (house or unit) and the same month, then looks at the evidence behind each score.

Two invented suburbs, A and B, run through this post. They are examples, not real places. Both score 58 overall, and they are very different places.

The weights are the ones PropRetire uses. The full list of inputs is in the suburb research guide. To see how suburbs are compared, open the suburb research page.

What makes two suburbs comparable?

A fair comparison keeps to houses or to units. A house suburb and a unit suburb answer different questions, so mixing them tells you little. It uses the same month too.

PropRetire compares up to four suburbs of one home type at a time.

Each score ranks a suburb against its state's peers, for that home type and month. A suburb in Victoria and a suburb in Queensland are each ranked in their own state.

The two scores are not on one national scale, so a 62 in one is not a 62 in the other.

Each figure has its own period. A sales count for one month and a count for twelve months are different measures. So are today's asking rents and rent growth over five years.

What evidence belongs beside the score?

A score sums up the evidence. Read the evidence too. PropRetire shows ranges for recorded sale prices and asking rents, with vacancy, time to sell and stock on the market.

• Rent: asking rents, vacancy and the number of rental listings. Rental market also counts yield, a year's rent as a share of the property's value, and the share of households that rent, so it is not a plain measure of demand.
• Selling: the typical time to sell, sales counts and stock on the market. Liquidity combines these. It does not say how fast any one property would sell.
• Price: recorded sale-price changes, from real sales, and modelled value changes, which are estimates for homes that did not sell. They are different things. Growth counts both, over different periods, and neither is a forecast.

Some things sit outside PropRetire's score: local jobs, new housing supply, flood or fire risk and works that are planned but not funded. Those come from council and state sources.

What do the five dimensions mean?

Each dimension scores one part of the evidence. A higher score means a stronger reading on that measure, and for Risk signals it means fewer warning signs. It does not mean a better investment.

• Growth: past changes in value and sale prices, and changes in sales activity.
• Rental market: yield, rents, listings, renters and vacancy.
• Liquidity: selling and listing activity, including time on the market.
• Stability: how uncertain the model's values are, how consistent repeat-sale prices are, and who lives there (the share of owner-occupiers and renters, and how long owners hold).
• Risk signals: scores higher when fewer warning signs show.

Growth and Rental market each carry 25% of the overall score, Liquidity 20%, and Stability and Risk signals 15% each. Some inputs count in two dimensions. Low vacancy helps both Rental market and Risk signals, so the five scores are not five separate checks.

Can two different suburbs have the same score?

Yes. Take two invented suburbs, A and B. A scores 68 for Growth, 50 for Rental market, 60 for Liquidity, 40 for Stability and 70 for Risk signals.

B scores 40, 70, 55, 70 and 60. Weighted as above, both come to 58.

A's Growth score is higher than B's. B's Rental market and Stability scores are higher than A's. The equal totals hide that, and they say nothing about which suburb would do better.

How much evidence sits behind a score?

Each dimension is built from inputs. It gets a score when the inputs that have evidence carry at least half of that dimension's weight. The overall score is full when all five dimensions score.

That does not mean every input had evidence. It is partial when three or four score and together carry at least 60% of the overall weight. Otherwise there is no overall score.

Missing evidence never counts as zero.

Say A had too little evidence for Stability and Risk signals. The other three carry 70% of the weight, so the score would be partial, and it would work out at 59 after rounding. It rose from 58 because two dimensions were left out, not because the suburb improved.

That 59 cannot fairly be set beside B's full 58, since the two rest on different evidence.

Read the completeness label and Data behind the score. That label reads Limited when the inputs with evidence carry under half the weight, and Moderate from half to under three quarters. Check the date too.

When the market evidence is out of date, PropRetire shows no score until the next release. A missing score is a gap, not a low score.

What still needs checking for your own property?

A suburb score does not tell you about one property: its condition, comparable sales, contract, planning controls, insurance, loan, rent or tax.

These records answer what a score cannot.

• Your lease and rent statements, against the area's asking rents. Rent received is a different figure from rent advertised.
• Your bills and loan repayments, which set against the rent show what is left. What is left of the rent after costs? shows the sum.
• Inspection notes and comparable sales for an address.
• Hazard maps, council planning information and insurance terms, from their own sources.

PropRetire shows each suburb's score, its five dimensions and the data behind it, and the suburb research guide explains how to read them. A comparison shows where two suburbs differ, and you decide what that means for you.