Suburb Research
What Makes a Good Investment Suburb? — The 2 Signals That Survived Backtesting
Ask any buyers agent what makes a good investment suburb and you’ll get a list that sounds thorough: proximity to schools, transport infrastructure, employment hubs, population growth projections, planned development pipelines. It’s logical. It also largely failed when we tested it against real outcomes.
We backtested 78 Australian suburbs — 28 that boomed, 50 that didn’t — and checked which characteristics actually separated the two groups. Most of the standard checklist failed. Two signals survived. Here’s what the data found.
The Standard Checklist
Property podcasts, buyers agents, and investment guides tend to converge on the same characteristics when describing a good investment suburb. The list typically includes some combination of the following:
The conventional checklist
- •Population growth and inbound migration
- •Infrastructure investment (transport, hospitals, major projects)
- •Building approvals and new supply pipeline
- •Employment driver proximity
- •Rental yield
- •Days on market
- •Vendor discounting levels
- •Distance to CBD or major employment node
Some of these are useful. Most aren’t — or at least, aren’t in the way they’re usually applied. The problem isn’t that the logic is wrong. It’s that these factors operate at the wrong scale and on the wrong timeline for suburb-level investment decisions.
What Backtesting Found: Three Failures
When we built an early formula around the three most commonly cited “growth driver” signals — population, infrastructure, and building approvals — here’s what the backtest returned:
55% accuracy across the tested suburbs. A coin flip. Here’s why each of the three big signals failed:
Population growth
Sounds logical — more people means more housing demand. But population data is coarse and slow. At suburb level, over a 1–3 year investment window, it had near-zero ability to distinguish which suburbs would outperform. Growth in a suburb’s population might take 5–10 years to translate into meaningful price movement, long after a boom cycle has started and ended.
Infrastructure spending
New train lines, hospital expansions, motorway upgrades — these stories drive enormous suburb speculation. But infrastructure affects whole corridors over decades, not individual suburbs on the timeline most investors actually work with. In backtesting, announced projects failed to discriminate boom suburbs from controls. The suburb next to a new station may boom. It may not. The infrastructure project alone doesn’t tell you which.
Building approvals
The theory: approval surges signal incoming supply pressure. Or alternatively, that high approval demand confirms buyer appetite. In practice, approval data is difficult to obtain at suburb level from free public sources, and it failed as a signal in backtesting regardless. State-level approval trends give you macro context, not suburb selection signals.
Key finding
Population growth, infrastructure spending, and building approvals are interesting macro context. But they have near-zero discriminating power at the suburb level over investment-relevant time horizons. The formula that used all three delivered 55% accuracy — statistically indistinguishable from guessing.
Want to know which signals actually work?
BoomAU scores 393 suburbs fortnightly using only the criteria that passed backtesting. Join the wishlist.
The Two Signals That Survived
After eliminating what failed, two characteristics survived every test we ran. Both are measurable with free data. Neither requires specialist access or forecasting.
Signal 1: Affordability Headroom
How a suburb’s median price sits relative to its capital city median. That’s it. Below the city median — outperform. Above 1.5× the city median — underperform. The effect is monotonic: the further below the city median, the stronger the expected outperformance. It was the only signal that reliably discriminated between suburbs within any given market period — not just across different years when the entire market moved.
The result that made this concrete: every single boom in the 78-suburb backtest was led by suburbs priced well below their city median. Not most. Every one. The suburbs that had already stretched to 1.5× or beyond the city median consistently underperformed regardless of their population trends, proximity to infrastructure, or past growth trajectory.
Every boom was below the city median
In 28 booming suburbs tested against 50 controls, the affordability gap was the clearest structural difference. Suburbs priced well below the city median had room to grow toward it. Suburbs already at a premium didn’t.
Mean reversion is the underlying mechanic. Markets exert gravitational pull on outliers. Cheap suburbs get discovered as buyers are priced out of adjacent markets. Expensive suburbs stagnate as that same buyer pool chases better value. The affordability signal is really just a structured way of measuring where in that cycle a suburb sits.
Signal 2: Boom Detection Timing
The second signal isn’t a prediction — it’s a detection. Rather than trying to forecast which suburbs will boom before they start, it identifies whether a suburb is currently in an active boom and, critically, how early in that boom it is.
The insight behind this approach: booms are multi-year events. Catching one 6–12 months after it starts still captures 60–85% of the total gains. With dramatically higher confidence than any predictive approach we tested. Detection says “is this suburb booming right now and does it have runway left?” — not “will it boom?”
Early-stage detection is measured by how much of the suburb’s affordability gap has already been consumed. A suburb that was priced 30% below the city median and has risen 5% is early. A suburb that was 30% below and has risen 25% is late — most of the headroom has closed. Same detection signal, very different timing.
Why detection beats prediction
Prediction tries to identify booms before the market signals them. At suburb level with publicly available data, this didn’t work — it produced coin-flip accuracy. Detection waits for the market to signal a boom and catches it early enough to capture most of the gains. The confidence improvement is substantial.
The 5-Component Detection Formula
Both survival signals are built into a five-component scoring formula. Affordability headroom has an explicit weight. The detection components — momentum, growth strength, and tightness — carry the rest.
| Component | Weight | What it measures |
|---|---|---|
| Momentum | 0.30 | Price growth acceleration |
| Growth Strength | 0.25 | Annual growth scored directly |
| Tightness | 0.20 | DOM + vacancy rate |
| Sustainability | 0.15 | Rental yield + vacancy trend |
| Headroom | 0.10 | Price relative to capital city median |
The formula produces a score from 0–100. A score above 80 signals an active boom. 65–79 flags an early boom. 50–64 is a suburb warming up. Below 50, the conditions for a boom aren’t present.
85.7% accuracy. Zero false positives. The 20.2-point separation gap between genuine booms and false signals means the formula doesn’t just reach the right conclusion — it does so with margin. When a suburb is booming, its score is well above the threshold. When it isn’t, it’s well below.
BoomAU scores 393 suburbs on exactly these 5 criteria
Fortnightly Strong Buy / Buy / Watch / Pass labels, filtered by your budget band. Join the wishlist.
The Hard Filters: Before Scoring Even Starts
Not every suburb gets a score. Before entering the formula, a suburb must pass four hard filters. Think of these as the minimum conditions for a boom to be detectable:
- Annual growth≥ 5%
- Days on market≤ 45 days
- Vacancy rate≤ 2%
- Median price≤ $800,000
The $800K price cap isn’t arbitrary. Every boom in the backtest was in a suburb priced well below the capital city median. Once a suburb crosses $800K it typically carries little to no affordability headroom — and affordability headroom is the strongest single signal the data found. Above $800K, the formula’s detection logic is operating in territory where the key precondition is absent.
A suburb that fails any filter gets no score. Not because it’s necessarily a bad suburb, but because the conditions for the detection logic to work aren’t present. Of 8,417 suburbs swept nationally, 393 currently pass all four filters and receive a score.
One Caution: Thin Markets
Days on market is a powerful tightness signal — but it has a sample size problem in low-transaction suburbs. A suburb that sells fewer than around 30 homes a year produces an unreliable DOM median. One fast sale can drag the figure down to 10 days. One slow listing can push it to 150. Below roughly 15 annual sales, the DOM number isn’t usable.
This is a practical implication, not an edge case. Many affordable regional suburbs sit below that transaction threshold. The formula produces a number, but thin-market results carry less conviction than results from suburbs with 50+ annual sales. Volume matters when reading these numbers yourself — and it’s why systematic filtering across thousands of suburbs is more reliable than reading individual suburb profiles.
How the Tiers Discriminate
Beyond detecting individual booms, the formula produces tiered labels that discriminate between suburbs even within the broader market cycle. The walk-forward backtest covered 12,360 postcode-months, with no lookahead. Excess return is each suburb’s 12-month growth minus the market median growth in the same period — the tide is cancelled out.
| Tier | Excess return | Beat market | n |
|---|---|---|---|
| Strong Buy | +7.5pp | 71% | 2,103 |
| Buy | +1.3pp | 55% | 3,349 |
| Watch | −0.7pp | 47% | 5,788 |
| Pass | −6.4pp | 28% | 1,120 |
Walk-forward backtest, 12,360 postcode-months, 2012–2026. No lookahead. Excess return = suburb 12-month growth minus market median growth. Full methodology →
Perfectly monotonic. Every step down a tier corresponds to worse outcomes on both measures: lower excess return and a lower proportion beating the market. The spread between Strong Buy and Pass is 13.9 percentage points of excess return. Strong Buy suburbs beat the market 71% of the time. Pass suburbs beat it only 28% of the time.
This is the same asset class. Australian residential property. The difference is suburb selection. A Strong Buy suburb and a Pass suburb may sit 40 minutes from each other. Over the same 12-month period, with the same national market conditions, their outcomes are structurally different.
The Full Scorecard: What Works, What Doesn’t
Every metric that featured in any version of the formula, and its verdict:
| Characteristic | Verdict | Why |
|---|---|---|
| Affordability headroom | WORKS | Only cross-suburb ranking signal that survived tide cancellation. Every boom was below the city median. |
| Boom timing (detection) | WORKS | 85.7% accuracy, 0% false positives on 78-suburb backtest. |
| Days on market | WORKS (in detection) | Effective tightness signal within the detection formula. Unreliable below ~30 annual sales. |
| Vacancy rate | WORKS (in detection) | Low vacancy confirms demand outstripping available rental supply. |
| Annual growth rate | PARTIAL | Works for detection (is it booming?). Does not predict which suburb outperforms within a period. |
| Rental yield | PARTIAL | Useful sustainability signal (0.15 weight). Not a standalone selection signal. |
| Population growth | FAILED | Too slow-moving and coarse for suburb-level timing over 1–3 year windows. |
| Infrastructure spending | FAILED | Affects corridors over decades, not suburbs over investment-relevant horizons. |
| Building approvals | FAILED | State-level data, weak suburb-level signal. Difficult to source at postcode level. |
| Vendor discount | FAILED | Data barely exists at suburb level from free sources. Cannot be used reliably. |
| Past 5-year momentum | FAILED | Mean reversion dominates. Past outperformers tend to underperform going forward. |
How to Apply This Yourself
Both signals are checkable with free, publicly available sources. You don’t need specialist access or proprietary data.
1. Check affordability headroom
Domain and CoreLogic publish capital city median house prices quarterly. Look up the median for the relevant capital city, then compare it to the suburb you’re evaluating. If the suburb sits below the city median, it has headroom. If it’s already above 1.5× the city median, the backtest says it’s less likely to outperform. This single check removes most overpriced suburbs from consideration immediately.
2. Check the boom conditions
YIP (yourinvestmentpropertymag.com.au) provides annual growth, days on market, and yield data at suburb level, backed by CoreLogic data. SQM Research provides free postcode-level vacancy rate charts going back 16 years. If a suburb shows annual growth above 5%, days on market under 45, and vacancy below 2%, you’re looking at the entry conditions for a detected boom. The earlier the suburb is in that trajectory — before the affordability gap closes — the more runway remains.
That’s the honest version. Two signals. Both free to check. The hard part is scale. Checking manually for a handful of suburbs is feasible. Checking systematically across thousands of suburbs every fortnight — catching booms within weeks of the signals appearing, filtering out thin-market noise where the numbers can’t be trusted — is what BoomAU automates.
Currently 393 suburbs pass all four hard filters and receive a score. Of those, 35 are under $400K, 149 are under $600K, and 204 are under $800K. The full backtest methodology and tier discrimination results are published on our proof page. No gating, no email required.
Join the Wishlist
We'll email you when BoomAU launches — starting with the budget range you care about.
Be first in line
- ✓Fortnightly Strong / Good / Fair / Weak signal labels per suburb
- ✓Filtered to your budget band
- ✓Built on a backtest of 12,360 postcode-months