This page is under development — content, data and layout may change.
FlatHuntResearch & Market Intelligence
Berlin · EuropeUpdated dailyFree public resource
FlatHunt Research · Methodology

Methodology and publication rules

How FlatHunt Research turns listing observations into published figures, and which checks a number must pass before it appears.

Methodology version: fh-observatory-v4Last data: 20 November 2026
01
Method

How figures are produced

What is measured

Advertised monthly asking rents of furnished and medium-term rental listings observed on partner platforms. These are offers, not signed contract rents, and not a census of every home.

Observation day

Each daily figure is the stock of listings at the end of the day (00:00 UTC). Monthly figures are the stock at the end of the month’s last observed day. A listing a platform omits for less than six hours still counts as listed.

Medians, not averages

Rents are summarised with medians and the middle 50% (quartiles), so a few very expensive listings do not dominate.

Like-for-like comparison

Cities list different homes (rooms, studios, large houses). Compare a segment (room, studio, 1 bedroom) or apartment rent per m², not the all-listings median.

Currencies

Each listing keeps the currency its platform quotes. Figures are shown in the city’s currency and converted to EUR at the European Central Bank reference rate of the observation day.

Plausibility checks

Rents outside 100–20,000 EUR-equivalent per month and per-m² rents outside 2–200 EUR are excluded as data errors. Per-m² rent is not published for rooms.

Coverage labels

“Single platform”: every listing in the city comes from one platform. “One platform dominates”: more than 70% from one platform. Figures then describe that platform’s supply.

Comparable history

Time series use platforms observed for at least 90 days, and chain-linked indices compare each month only with platforms present in both months, so adding a platform does not look like market growth.

Truncated sources

Platforms whose catalogue reaches FlatHunt only partly (for example limited to a price range) are excluded from market figures.

Duplicates

The same home is often listed on several platforms. Since 20 November 2026 figures count homes: listings confirmed as the same home by matching photos (plus location, size and rent) count once, priced by the listing that has been online longest, and a home is new or removed only when it appears on or leaves every platform. A home holds at most one listing per platform, so identical units of one building are not merged. Earlier days count listings; supply and flow series are therefore compared within one basis, and the chain-linked indices, built per platform, are unaffected.

02
Publication rules

What is published

FlatHunt stores more than it publishes. A figure appears only when it meets these minimums; otherwise it is shown as a dash or the page is held back.

FigureMinimum
City page50 priced listings on the latest day; 100 when one platform supplies all listings.
Segment30 priced listings (e.g. studios, 40–60 m² apartments).
Shares30 listings with a known value (deposit, Anmeldung, pets, price history); 50 with an amenity list.
Districts30 priced listings of a home type.
Activity and time on marketAfter 90 days of tracking a city, over a full 28-day window.
HistoryFrom 2 months of observations.
03
Data

Downloads and API

All published cities (CSV) Publication index (JSON)

City reports: https://api.flat-hunt.com/api/public/research/cities/{cityId}/report. Please cite “FlatHunt Research” with the observation date. Data use terms