How we compute the numbers

What each number means

Median. The value that sits exactly in the middle when you sort every listing in an area. One very expensive penthouse does not move it. We never publish averages.

US$/m². Listed price divided by the area declared by the agent. Storage units and parking spaces do not count as area. It is computed only over listings that declare an area, which are fewer than the total, which is why we show both counts.

Active listings. Adverts that were visible on Suelo during the month. They are not sales or closed transactions.

New listings. Adverts that appeared for the first time during the month.

Monthly change. Change in the median against the previous month, in the same area and the same property type. Published only when the area has at least 30 listings in each of the two months.

Area. The finest level we report on: the neighbourhood or parish. When there are not enough listings, we move up to the municipality and then to the state, and we tell you which level the number came from.

From listing to data point

1. Collection. We take the listings active on Suelo during the month. Most come from public property portals we crawl, and the rest from agents who list directly on Suelo. We do not verify the advertiser's identity in order to include their listing in the calculation.

2. Normalisation. Each listing is anchored to a single geographic key (ve / state / municipality / neighbourhood), the amount is converted to dollars and the area to square metres.

3. Deduplication. The same unit listed by several agents is counted once, matched on address, area and price.

4. Aggregation. We compute the median and the quartiles by area and property type.

5. Publication. The report goes out with the same numbers you see in any chart on the site, and it is then frozen: a closed month is not recomputed.

The two thresholds

No area appears with fewer than 5 listings. Below that, the median depends on which adverts happened to be up that day and does not describe a market.

No monthly change is published with fewer than 30 listings in each of the two months. That threshold is far higher because a change is the difference of two medians and the noise of both adds up. We measured the dispersion inside each area: at 5 to 7 listings, month-to-month noise reaches ±57%, enough to invent a 30% rise where nothing happened. At 30 or more it falls to ±16%.

This is why the report does not yet publish a "biggest movers" ranking. Sorting by largest change selects precisely the noisiest areas. It will come back once we have enough months to compute change over the same properties rather than over two different groups.

What we leave out

Areas with fewer than 5 listings. They roll up into the municipality; they are not discarded.

Listings with no declared area. They are excluded from the US$/m² figure but included in the median price.

Listings with no price. Excluded from every price calculation.

Cross-portal duplicates. Counted once.

What this data is not

It is not a closing price. We publish asking prices: what the seller wants, not what the buyer paid. In Venezuela the closing price is not public, so no index, ours included, can state real transactions.

It is not a valuation. For one specific unit, the state of the building, the floor, the view and the services matter more than the median of its neighbourhood. Use it to orient yourself, not to set your price.

It is not a census of the market. It is a census of what gets listed on the portals we crawl. Where our coverage is thinner, the median describes a smaller slice of the real market. That is why we publish the listing count next to every number.

It does not include days on market. We can measure when a listing appears, but we cannot yet reliably tell a sold property from one the agent delisted and relisted. We have been recording the data since August 2026 and will publish it when it is trustworthy, not before.

It does not separate furnished from unfurnished. The field exists but is empty in practice across the rental listings we receive.

You may use these numbers

Any figure published in these reports can be reproduced in a news article, a thesis, a market report or a presentation. All we ask is that you credit Suelo and link to that month's report.

Suelo (2026). Monthly Venezuelan property market report — July 2026. suelo.casa/datos

We do not yet publish a downloadable file with the full series. If you are working on research and need the historical series, write to us and we will look at it case by case.