Methodology
How RealData measures every Thai condo. Sources, formulas, refresh cadence, and what we deliberately don't do.
Founder
Shin Yunmin (신윤민)
Data architect, 12+ years. Korean founder based in Bangkok, building products for the Thai market.
Why RealData exists: “Top 10 Bangkok condos for foreigners” videos on YouTube are almost always paid placements by developer sales agents. Influencers walk a showroom, mention “great yield potential,” never mention the building’s sales velocity, the district’s flood risk, or the realistic rent. So I built the opposite: 1,894 buildings, 23,552 units, weekly portal sweeps, BOT mortgage rates, BMA flood layers. Buy on evidence, not influencers.
What we do
We measure 1,700+ condos across Bangkok, Phuket, Chiang Mai, Pattaya, Hua Hin, and Chonburi continuously. Every building gets a Bubble Index, livability score, and (in Bangkok) a flood-risk level — independent of the listing source's marketing copy. No referral fees, no ad placements, no influencer deals.
Where the numbers come from
- Listings & price
- hipflat.co.th, weekly re-crawl. Per-unit listings (sale + rent), area, bedrooms, publisher.
- Flood risk (Bangkok)
- BMA Drainage Department + JICA reports + 2011 great-flood inundation records. District (khet) baseline, not coordinate-level.
- Transit & amenities
- OpenStreetMap via Overpass API. BTS/MRT stations + hospitals/schools/supermarkets within 1km.
- District boundaries
- OSM admin_level=6 polygons. Used for choropleth + per-district aggregations.
How we score
- Bubble Index
- (building's median price-per-sqm ÷ district median price-per-sqm) × 100. 100 = at market. 200 = double the local average. <80 = underpriced. >130 = bubble suspect. Districts with fewer than 5 sampled buildings excluded (small-sample noise).
- Livability Score
- Weighted aggregate of BTS/MRT distance, hospitals/schools/supermarkets within 1km, and total amenities. 0–100 scale.
- Super Value flag
- Underpriced (Bubble Index < 90) AND top-quartile livability.
- Flood Risk (0–5)
- 0 = none observed, 1 = very low (central elevated), 2 = occasional puddling, 3 = neighborhood flooding common, 4 = waist-deep recurring, 5 = severe full-area inundation.
What we deliberately don't do
- ✗We don't broker listings. Listing clicks go to the original hipflat page.
- ✗We don't take referral fees from agents or developers.
- ✗We don't run banner ads. Marketing-vs-Reality slots are paid promotion — clearly labeled.
- ✗We don't accept influencer placement or 'sponsored content' that shapes our data.
- ✗We don't publish bubble_index values above 10× district median — those are data-quality issues, not real prices.
Refresh cadence
- Listings & price
- Weekly hipflat re-crawl
- Flood risk
- Annual (after BMA monsoon report)
- Transit & amenities
- Quarterly review on OSM change detection
Each building page surfaces its measurement timestamp.
What we can't see (yet)
- !Block-level flood polygons — BMA does not publish them. We use district-level baselines.
- !Foreigner-quota availability per building — varies by project office. Confirm before purchase.
- !Off-plan / pre-construction pipeline — we only measure standing inventory listed on hipflat.
- !Resale velocity — we have median sale price, but not days-on-market or sale-through ratio yet.
FAQ
Why should I trust RealData over an agent?▾
Because we publish our methodology, our sources, and our refresh cadence — and we don't get paid by listings. An agent's incentive is closing the sale; ours is being correct. Verify any number we publish against the underlying hipflat page (linked from every condo report).
Is RealData affiliated with hipflat?▾
No. We crawl hipflat's public pages within their robots.txt and rate-limit policies, the same way Google or Bing does. We attribute every listing back to its original hipflat URL. We have no revenue share or partnership with them.
Why is the Bubble Index missing for some buildings?▾
Three reasons: (1) the building has no listings on hipflat yet, (2) its district has fewer than 5 sampled buildings (peer baseline too thin), or (3) the computed value was implausible (>10× district median, treated as a data-quality issue and dropped).
Why is flood risk only available in Bangkok?▾
BMA (Bangkok Metropolitan Administration) is the only authority publishing district-level monsoon flood records. Phuket, Chiang Mai, etc. have provincial-level flood data, but not the per-district granularity needed to score individual buildings. We're researching alternative data sources for those markets.