Property research: sources and limitations
I separate advertised prices, official new-build sales statistics and developer offers. They describe different markets and should not be treated as interchangeable.
Updated - Documented historical listing-research evidence gaps and limitations.
Spring 2026 listing estimates
The Where to buy guide reports 13,647 apartment listings across seven markets. The seven published subtotals add to that number, but this checks arithmetic only, not the underlying sample.
The retained research material does not include the original listing exports, identifiable portal sources, exact collection dates, duplicate-removal rules or reproducible calculations. I cannot independently reproduce the sample count, price medians, neighbourhood ranges, listing-age shares or rental yields from those records. I do not describe these figures as verified.
Spring 2026 is the period stated in the original guide, not a confirmed collection window. A later page update does not mean the market data was refreshed. These figures remain historical estimates pending source recovery or a new analysis.
The September asking-price snapshot
The asking-price table is a separate snapshot labelled 13 September 2026. Its retained town-level figures are not a live feed and do not substantiate the Spring sample. The original source records and calculation rules are not retained alongside that table, so its medians, percentiles, counts, listing ages and yields have not been independently reproduced.
Asking prices are not sold prices
An advertised price is a seller's request, not a completed transaction. I do not infer an achieved sale price or a standard negotiation discount from it. The MONSTAT page covers official contracted new-build statistics and identifies its source separately. It does not validate the historical listing estimates or describe every resale.
Listing age and rental yield
Days advertised are not days to a completed sale or tenancy. An advert may disappear without a transaction, while relisting or duplicate adverts may change the apparent age. Without a documented first-observed date and property-matching rules, I cannot use listing age as a verified measure of resale liquidity.
A gross rental yield illustration normally compares annual rent with purchase price before costs. That general definition does not establish how the published figures were calculated. Their rent inputs, matching rules, occupancy assumptions and denominators are not documented. They are not guaranteed returns and do not establish net income after vacancy, management, charges and tax.
What a reproducible update needs
- Named sources, saved source records and exact collection dates.
- Property-type and geographic eligibility, with sample sizes before and after cleaning.
- Rules for duplicate adverts, multi-agent listings and relisted properties.
- Area and currency normalization, missing-data handling and median or percentile calculations.
- First-observed dates for listing age and documented rental-yield inputs.
These are requirements for future analysis, not claims that those steps were performed on the historical samples.
Developer facts and my assessment
Developer pricing is separate from listing-market estimates. Use the dated source and current property page for each offer; a grouped rate does not establish individual availability. My location recommendations are professional assessments, not measured investment outcomes.