SG Property Index is the first platform in Singapore that verifies whether property agents actually know the market. The system is simple: agents submit prices, HDB publishes official data, and we compare the two.
When you buy or sell an HDB flat, the single most important number is the transacted price. But HDB publishes this data two to three months after each transaction. That means buyers and sellers are making decisions worth $400,000 to $800,000 based on data that is already outdated.
Property agents fill that gap. They tell you what similar flats recently sold for. The problem is that there is no way to verify whether those numbers are accurate. An agent can quote any price, and you have no independent way to check until HDB publishes the real figure months later.
CEA (the Council for Estate Agencies) tracks disciplinary actions and complaints. But nobody tracks whether agents price accurately. Until now.
After closing a deal, the agent logs into SG Property Index and submits the transacted price along with the property details: block, street, town, flat type, storey range, floor area, and transaction month.
Every agent must verify their CEA registration number before submitting. The submission is timestamped and locked after 24 hours. No edits, no backdating, no retroactive corrections.
The price shows up on the platform immediately, clearly labelled as an agent-reported figure pending official verification. Buyers can see near-real-time transaction data, but they also know it has not been confirmed yet.
Two to three months later, HDB releases the actual transacted prices on data.gov.sg. This is the ground truth. Every resale transaction in Singapore is recorded here.
Our matching engine compares each agent submission against the official HDB record. It matches on town, flat type, block, storey range, floor area, and transaction month. When a match is found, the accuracy is calculated as a percentage.
Only submissions made before the official data was published count towards an agent's score. You cannot game the system by submitting after the fact.
Each verified match contributes to the agent's overall trust score, which is a weighted average of all their verified predictions. Recent matches count more than older ones. The score and full accuracy history are public.
Each verified match gets a star rating based on how close the agent's submitted price was to the official figure:
| Rating | Accuracy | What It Means |
|---|---|---|
| ★★★★★ | Within $5K | Exact match |
| ★★★★ | Within 3% | Very close |
| ★★★ | Within 5% | Reasonable |
| ★★ | Within 10% | Off |
| ★ | >10% | Way off |
As agents build up verified transactions, they earn trust tiers:
The system is designed so that the only way to get a good score is to actually know the market.