Two Condos Launched In 2007. One Gained 62%. The Other Lost 26%.

Both were sold as good buys. Only one of them was.

Here is what separated them — and why we spent two years testing whether it could have been spotted in advance.

In 2007, two private condominiums launched in Singapore within months of each other.

The first was One-North Residences in Buona Vista.

Launch median: $881 per square foot.

Ten years later, units there were transacting at $1,427. A buyer who put in $1 million was sitting on roughly $1.62 million.

The second was Duchess Residences in Bukit Timah.

Launch median: $1,813 per square foot — more than double.

Ten years later: $1,342. That same $1 million was worth about $740,000.

One launch gained 62%. The other lost 26%. Same year. Same market. Same interest rates, same cooling measures, same everything.

I want to be careful here, because this is the point where most property articles start telling you they saw it coming. I didn't.

In 2007 I would have looked at Duchess Residences — freehold, Bukit Timah, low density, beautiful — and thought it was the safer of the two.

That is exactly the problem. It looked safer. It wasn't.

What was actually visible in 2007

Go back and look at what anyone could have checked on launch day, without a forecast, without insider information, without knowing anything about what the next decade held.

One-North Residences: 503 metres to the MRT. 405 units. A large project in a district the government was actively building out, surrounded by HDB flats whose owners would soon hit their five-year mark and start looking to upgrade.

Duchess Residences: 1.0 kilometre to the nearest MRT. 120 units. Almost no upgrader catchment nearby. And an entry price of $1,813 per square foot.

Every one of those facts was public in 2007. None of them required a crystal ball. The walk to the station is on a map. The unit count is in the brochure. The HDB blocks around the corner have completion dates you can look up.

The information was all there. Nobody had a way to weigh it.

That is the gap we set out to close.

Why we tested it backwards

Any framework can be made to look clever after the fact. If I pick two projects today, I already know how they turned out, and I can construct a story that makes my system look prescient. Everyone in this industry does it. It proves nothing.

So we did it the other way around.

We took 718 Singapore private condominium launches between 1995 and 2026 and back tested them.

For each one, we reconstructed a score using only information that existed on the day it launched — the walk to the station, the unit count, the lease remaining, the schools within a kilometre, the HDB flats approaching upgrade age, the land being released nearby, the rental market at the time, and whether it sat inside a designated growth area.

No hindsight. No adjusting the weights until the answer looked good.

Then we checked what each project actually transacted at ten years later, using URA data.

Every figure in this article has been computed thrice or more — from our own records, and from a panel of 619,093 raw URA transactions covering January 1995 to August 2026.

Only figures where testing control methods agreed are published. When data is not clean or sufficiently representative of the average, it is removed.

Duchess Residences came back at −26.0% on both. One-North at +62.0% and +61.4%.

What came out

Here is the part that matters.

We sorted all 718 launches into four grades based on their launch-day score, then took the median ten-year growth of each group:

Grade at launch	Median 10-year growth
Excellent (36/40)	+40.7%
Good (30/40)	+30.8%
Fair (24/40)	+22.5%
Poor (18/40)	+6.8%

Four grades. Four clearly separated outcomes. In the correct order.

I have shown this table to a few people who assumed it was too neat to be real, so let me say plainly what it is and is not.

It is not a promise that a well-graded condo will return 40%. That is a median across a large group over a specific historical window, and the next decade will not repeat the last one.

What it does tell you is that the 8 things we measure were, over thirty years and 718 projects, systematically related to what happened next. The ordering holds whether you measure raw growth or growth relative to other launches from the same year — which matters, because the second test strips out the effect of simply having bought in a good year.

And the gap that should hold your attention is not the top. It is the bottom.

A Poor-graded launch returned 6.8% over a decade. Not 6.8% a year. 6.8% in total, before you subtract stamp duty, legal fees, agent commission, property tax and ten years of interest.

In real terms, most of those owners went backwards.

The one that should worry you

In 2010, Seascape launched in Sentosa Cove at $2,683 per square foot.

Ten years later it was transacting at $2,047. A loss of 23.7%.

On our scale it scored 15 out of 40 — Poor. Seven of its eight pillars came in at two stars or below. No MRT within walking distance. No upgrader catchment. Thin rental demand.

Limited exit liquidity.

The single pillar it scored well on was remaining tenure, and a long lease cannot rescue a property that nobody can conveniently get to and few people are queuing to buy.

Every one of those seven weaknesses was visible in 2010.

I bring up Seascape rather than a more obscure example because nobody bought there by accident. These were considered purchases, by sophisticated buyers, in Singapore's most exclusive residential enclave. The buildings are genuinely beautiful. The marina is real. None of that was a lie.

It simply had very little to do with what the property would be worth in 2020.

The eight things we look at

There is no secret sauce here, which is deliberate.

MRT connectivity. Growth hotspots. Government land sales and enbloc activity nearby. Project size. Remaining tenure. School effect. MOP cluster — how many HDB households around you are approaching the point where they can sell and upgrade. Rental yield.

All eight are measurable. All eight are knowable before you sign anything. None of them requires me to predict interest rates or guess at the next cooling measure.

The weighting shifts depending on who is buying when required of a specific profile match - PrimeKey Profiles is what we call it. Another article on it soon.

A young family with a school-going child and an investor buying purely for yield are doing different jobs with the same money, and the same building can be excellent for one and poor for the other. But the inputs stay the same, and they stay checkable.

What this does not do

I would rather tell you the limits now than have you find them later.

It will not find you the biggest winner. Picking the top performer of a decade requires luck and the right entry price, and any framework claiming otherwise is selling something.

PrimeKey works best on new launches.

On resale, thousands of buyers have already spent years arguing about what a block is worth, and most of what we measure is already reflected in the asking price by the time you arrive.

It scores the property, not the price. This is the one that catches people.

A five-star building at a six-star price is a bad investment — Eight Riversuites and Parc Centros both scored 36/40 in 2012, and one returned 44% while the other returned 16%.

The difference was entirely what buyers paid on day one.

That is why we always have to run an entry-price check alongside the score.

And it is a research tool, not financial advice. It narrows a shortlist. It does not make the decision.

The point of all of this

In Navis, we did not build this to find you a winner. We built it because we have watched too many people put most of their savings into one property, in one location, for ten years, on the strength of a showflat and a feeling.

For someone holding four properties, a bad one is an annoyance. For a family selling their flat to upgrade, a bad one changes their retirement.

The 718 launches PrimeKey backtests say something quite specific about that risk. Graded Poor, roughly one in six lost money outright over ten years.

You do not need a system to tell you that One-North Residences worked out.

You needed one in 2007, when Duchess Residences looked like the safer buy.

If you are looking at something now, look for any Navis agent online and they will generate a report that runs through all eight factors — and show you where it is strong and where it is weak.

No obligation, and no recommendation to buy anything. 

Stuart Chng is the Managing Partner of Navis and Chief Agency District Director at Huttons and the co-creator of Navis Atlas and PrimeKey Analysis.

He adores music and can play a few instruments decently without upsetting his neighbours. When not doing so, he enjoys pillow fighting with his son and coming up with silly puns which barely amuses his wife. 

Professionally, he is a licensed real estate agent, avid investor in options, stocks and real estate, team leader, speaker and columnist for several property newsletters and blogs and is often quoted in media interviews on 938FM, Channel 8, PropertyReport, PropertyGuru and other publications.

Throughout his career, he has helped many clients grow their wealth through selecting great real estate investments and managing their portfolios actively. Read his clients' reviews here.

Stuart has also coached many top million dollar producing agents from top Singapore real estate agencies. Read his agents' reviews here.

*All figures from the PrimeKey 1995–2026 back-test on URA transacted data, verified against 619,093 deduped transactions. Past performance is not a guarantee of future results. PrimeKey Analysis is a research and shortlisting tool built by Navis, not financial advice.