Data provenance
Every number has a paper trail.
Datanah cleans public Malaysian property transaction records and turns them into searchable development, district and state statistics. This page documents what we keep, what we calculate and where the data falls short.
- 01ReceivePublished records
- 02StandardiseNames and fields
- 03CalculateMedians and trends
- 04PublishTraceable results
01 / Coverage
Recorded sales across Malaysia
Datanah covers recorded sub-sales from January 2021 onward. The source includes residential, commercial and industrial property transactions across Malaysia's 13 states and three federal territories.
State, district and mukim where supplied.
Development name, property type, road and tenure where supplied.
Recorded sale date, price and valued floor where supplied.
Main floor area and land area where supplied.
Datanah publishes recorded sale prices. Listing prices, formal valuations, rental records and forecasts sit outside this dataset.
02 / Source and terms
Source and permitted use
Datanah uses the government's published Open Sales Data. We maintain the cleaning rules, calculations and presentation used across this site. The source owner does not endorse Datanah.
Required attribution
Source: Open Sales Data published by the National Property Information Centre (NAPIC), Valuation and Property Services Department (JPPH), Government of Malaysia. Datanah independently cleans, standardises and analyses the records.
Data and information are subject to the Malaysian Government Open Data Terms of Use 1.0Read the Government Open Data Terms of Use 1.0
03 / Cleaning
The records Datanah accepts
We reject rows without a state, district, development, property type, transaction date or positive sale price. Accepted records pass through the same repeatable process on every release.
- 1Standardise names
We fix spacing and casing and expand a controlled set of unambiguous abbreviations. We keep the original development name with the stored transaction.
- 2Preserve genuine repeats
A repeatable record fingerprint prevents the same release from loading twice. An occurrence counter keeps separate units that share the same date and price.
- 3Complete the release
Batches load sequentially. After the final batch, Datanah rebuilds the development and monthly statistics and marks the release complete. Transaction pages may begin reflecting new rows while that processing is under way.
04 / Calculations
How the headline figures work
Median sale price
For a development headline, this is the middle price among sales in its assigned category. A development is residential when at least 30% of its sales are residential; otherwise commercial or industrial is chosen by whichever has more sales. Its total transaction count still includes every accepted category.
Price range
The 25th to 75th percentile contains the middle half of recorded prices included by the current page and filters.
RM per square foot
Datanah divides the sale price by main floor area when supplied. Land area is the fallback when floor area is missing.
Trends
Trend windows end at the latest month with records. We withhold comparisons when either period has too few eligible sales.
When floor area is missing, Datanah uses land area instead. The resulting RM psf figures share one field even though the two measurement bases do not mean the same thing. Treat rankings and comparisons across different property types with care.
05 / Limitations
Read the gaps with the numbers
A missing record does not prove that no sale occurred. Publication schedules, source revisions and incomplete fields can affect what appears on Datanah and when it appears.
- The source does not supply bedrooms, unit numbers or asking prices.
- Exact coordinates and reliable completion years are unavailable.
- Some records omit floor area, land area or tenure.
- New rows can appear while an import is still being completed.
- Datanah's statistics are research aids, not valuations or financial advice.
Inspect the evidence