Vendor Discount Property Meaning — And Why It Failed as a Suburb Signal

Vendor discount measures the gap between what a vendor originally listed their property for and what it actually sold for. If a home is listed at $600,000 and sells for $570,000, the vendor discount is 5%.

In theory, this is a real signal. A suburb where vendors consistently accept large discounts is a suburb where buyers hold the power — more supply than demand, slow absorption, sellers sitting on properties longer than they planned. A suburb where properties sell at or above asking is a suburb where buyers are competing. The direction of vendor discounting should tell you something about where the market balance sits.

The logic is sound. The problem is the data.

This is a practical constraint that affects almost every investor trying to do their own analysis. You might be able to look up the history of a specific property you’re considering buying. Aggregating that across every sale in a postcode, over 12 months, for 400 postcodes — with consistent methodology — is a different problem entirely.

BoomAU’s first suburb-scoring formula included vendor discount alongside six other inputs: inventory change, days on market, vacancy rate, population and migration data, infrastructure spending, and new supply from building approvals. The idea was to build a formula that could predictbooms before they started — identifying which suburbs were primed to outperform over the next 12–18 months.

Seven inputs. Months of research. 55% accuracy — which means the formula was essentially guessing. Adding more variables to the model didn’t make it better. It made it more confident while remaining wrong roughly half the time.

The core issue was that the formula was trying to predictbooms before they started. Infrastructure spending, population inflows, and vendor discounting patterns are slow-moving signals that operate at corridor or city level. They don’t tell you whether the suburb two streets over is about to see prices jump 20% in the next 18 months. At suburb granularity, over investment-relevant time horizons, none of these inputs held predictive power.

The pivot that changed everything was a question: what if we stop trying to predict booms before they start, and instead detect them once they’re already running?

The instinct against this is understandable. The whole point of analysis is to be early, right? But actual boom trajectories across Australian suburbs tell a different story. Booms are multi-year events. Getting in 6–12 months after one starts still captures 60–85% of the total gains — with dramatically more confidence than a prediction that might be right 55% of the time.

Detection requires different metrics than prediction. You’re not asking “which suburb will boom?” — you’re asking “is this suburb booming right now?” That question has answers you can find in free, reliable, regularly-updated data.

Notice what’s not on that list. Vendor discount isn’t there — not because the concept is wrong, but because the free data doesn’t exist at the granularity needed. Days on market serves a related purpose (measuring how fast buyers are absorbing stock), and the underlying data is freely available from YIP and Domain for thousands of Australian suburbs.

The current formula — version 2.3 — has five components, each weighted by how much predictive work it carries in the backtest.

Tightness — the component that carries days on market and vacancy rate — gets 20% of the score weight. That’s the slot where vendor discount would sit if the data existed reliably. The difference is that DOM and vacancy are freely available for thousands of Australian postcodes. Vendor discount at the same granularity is not.

The formula also applies hard filters before any suburb gets scored. A suburb must pass all four to be included in the results:

These filters cut the universe from thousands of Australian suburbs to the 393 that currently pass. They’re not arbitrary — the DOM and vacancy thresholds reflect the tightness conditions that showed up in every detected boom in the backtest.

The detection formula was validated against 78 suburbs — 28 that boomed and 50 controls that didn’t. The results were materially different from the prediction approach.

The 20.2-point separation gap is the number that matters most here. It means the formula doesn’t just get the right answer by a hair — booming suburbs score clearly above non-booming suburbs. There’s no muddy middle where a suburb sits at 62 and you’re guessing.

One finding from the backtest that shaped the hard filters: every single boom was led by a suburb priced well below its capital city median. Not one boom in the 78-suburb dataset came from a suburb sitting above 1.5 times the city median. The affordability threshold isn’t a preference — it’s what the historical data showed consistently.

There’s a legitimate critique of days on market as a suburb-level signal, and it applies even more forcefully to vendor discount: thin markets distort the numbers.

A suburb that sells fewer than roughly 30 homes a year produces a DOM median that’s just an echo of a handful of transactions. One fast sale from a motivated vendor — priced to move in a week — can pull the suburb median down to 10 days. One slow listing from an owner who’s in no hurry can push it to 150. Below about 15 annual sales, the number isn’t usable at all.

The same problem would apply to vendor discount, except worse: instead of needing 30 sales to get a reliable DOM median, you’d need 30 sales where the original list price is also reliably recorded — a higher bar that most free sources don’t clear.

BoomAU’s hard filter on median price (under $800K) partially addresses this by excluding the high-end, low-transaction suburbs where thin-market distortion is most common. But any investor using raw vendor discount numbers from a data service should be aware of this effect: the smaller the suburb, the less meaningful the aggregate figure.

The formula outputs a tier label — Strong Buy, Buy, Watch, or Pass — for each suburb. Those labels were validated in a walk-forward backtest across 12,360 postcode-months.

Walk-forward backtest, 12,360 postcode-months. Excess return = suburb 12-month growth minus market median growth. Full methodology →

The tiers are perfectly monotonic. Strong Buy outperforms Buy, Buy outperforms Watch, Watch outperforms Pass — at every step. The spread between Strong Buy and Pass is 13.9 percentage points of excess return per year. That’s the gap between a formula built on backtested signals and one built on signals that sounded good but didn’t survive testing.

Vendor discount data isn’t freely available at suburb level, but the signals that replaced it are. Here’s what you can check on public sources:

These three checks — DOM, vacancy, affordability — are what the formula is built on. The hard part isn’t knowing which signals matter. It’s applying them consistently across hundreds of suburbs, updated fortnightly, filtered to your budget, without missing the booms that start while you’re focused on something else.

Full backtest methodology and the 78-suburb validation results are published on our proof page. No gating, no email required.