How these numbers are made, and what they are not

A. The source

Every town figure comes from live public listings on ikman.lk, Sri Lanka’s largest property marketplace, pulled on 2 November 2026. Each sample ad links out to the original listing so you can check any number yourself. Listings expire and sell, so this is a dated snapshot, not a live feed.

B. Asking ≠ selling

These are the prices sellers ask, not the prices buyers close at. In Sri Lankan practice, final prices typically settle below asking. Treat each median as the top of the negotiation, not the signed number.

C. The cleaning

Ads with missing prices are dropped. Obvious junk is filtered by hard bounds (a Rs 180 “house” is not a house), then every town-metric group is trimmed to its 5th–95th percentile band before the median is taken. Ads whose title names a different town than the tagged location are excluded outright — 1,572 of them, because sellers mislabel locations to catch searches. A median of 30 real ads survives a liar; a mean does not.

D. The trend lines

The index chart is the Central Bank of Sri Lanka’s Real Estate Asking Price Indices for Colombo district (2019=100), from its quarterly Real Estate Market Analysis reports. Where CBSL publishes levels we use them verbatim; where it publishes year-on-year changes, earlier levels are reconstructed backwards one month at a time. The 2019–2026 average-price cards are LankaPropertyWeb’s published quarterly averages. Our own daily town tape starts 25 August 2026 — before that, no town-level history exists on this site, and we say so rather than invent it.

E. The arithmetic

Per-perch rates come straight from per-perch priced ads, or from total price ÷ stated perches. Per-square-foot rates are asking price ÷ stated floor area. Yields are median annual rent ÷ median asking price in the same town. Nothing is modelled, fitted or forecast — it is division.

F. Sample sizes, always shown

Every number on this site carries its n. A median from 4 listings is a hint; from 40, a market. League tables require 8+. Where a town is thin, the calculator says so and falls back to its district — and tells you it did.

G. Signals, not advice

The opportunity cards combine published index momentum with today’s listing medians. They point at where to dig; they are not a recommendation to buy anything. Verify every figure against the linked source before money moves.

The accuracy checks are in the data sources and accuracy report (PDF).