The price displayed on a real estate listing and the price actually paid in front of the notary are two distinct pieces of information. To estimate the value of a house near you, it is the latter that matters. Since the opening of land data to the public, several tools allow access to these amounts. Their use requires some precautions that public platforms do not always detail.
Geographical and Temporal Limits of DVF Data
The Demande de valeurs foncières (DVF) file, published by the government, lists real estate transactions recorded by tax services. Its depth covers the last five years, with a semi-annual update (April and October).
The coverage is not uniform across the territory. Alsace-Moselle and Mayotte are not included in the DVF database, due to specific land regimes. In Paris, Lyon, and Marseille, the data is exploited at the district level, which limits precision by neighborhood.
Before comparing a price from DVF to a current listing, it is essential to check the year of the transaction and the date of the last transfer. A sale recorded three years ago in an area where prices have changed significantly does not provide a reliable benchmark for setting a selling price today. When looking to find a site for sold house prices, this verification of data freshness remains the first useful step.
Price per m² by Municipality and Individualized Estimation: What DVF Does Not Say
Platforms that utilize DVF often display a median price per m² by municipality or postal code. This indicator provides a market trend, not an estimate of your property.
Two neighboring houses can have very different values despite having the same local average price. The general condition, the size of the land, the exposure, the energy performance diagnosis, or the presence of a view can alter the actual value without these criteria appearing in the DVF data.
Specifically, DVF provides:
- The selling price and the date of the transaction, as recorded in the notarial deed
- A brief description of the property (type, number of rooms, area)
- The median selling price per m² at the national, regional, and municipal levels
What is systematically missing: the floor for an apartment, the interior condition, the renovations made, the DPE, the quality of the immediate environment. The available data does not allow for conclusions about the value of a specific property based on a single aggregated indicator.
Notarial Databases: A Complementary Angle to the DVF File
Notaries have their own databases, fed by the deeds they sign. These databases are not identical to DVF: they include qualitative information collected during the drafting of the deed.
An estimate made by a notary relies on comparable sales from notarial databases and a physical visit to the property. This dual approach allows for adjusting the gross price per m² based on the actual characteristics of the property. The site notaires.fr provides access to some of this data, with price maps by sector.
Field feedback varies on this point: some owners believe that an online estimate is sufficient, while others notice a significant gap between the price suggested by an automated tool and the one validated after a visit. The difference often lies in elements that are invisible in a database (nuisances, easements, quality of the co-ownership for an apartment).
Seasonality of Real Estate Sales: An Underutilized Parameter
Beyond price, the timing of the sale influences the number of comparable transactions available. A dataset published on data.gouv.fr analyzes the temporal distribution of sales by municipality and department, with data going back to 2014 depending on the areas.
If you consult DVF in January for a suburban neighborhood where most sales occur between April and July, the number of recent references will be mechanically low. The median price calculated on two or three transactions does not have the same statistical reliability as an indicator based on several dozen sales.
This local seasonality also helps choose the right time to sell. An active local market in spring generates more fresh comparables and often a more pronounced competition among buyers.
Cross-Referencing Sources to Validate a Selling Price Estimate
No single source is sufficient to set a realistic selling price. The most robust approach combines several levels:
- DVF to identify recent transactions in a nearby radius and verify local trends
- Notarial databases for qualitative comparisons that incorporate the actual condition of properties
- A visit or professional opinion (notary, real estate agent) to adjust the price to the specifics of the property
- Analysis of local seasonality to assess the statistical relevance of the consulted data
The median price per m² serves as a starting point, not a selling price. The gap between this indicator and the reality of a transaction can reach significant proportions depending on the condition of the property, its precise location, and the timing of the sale.
A plot of land or an apartment located in the same municipality as a house does not follow the same price dynamics. Verifying that the properties compared in DVF correspond to the same type of property remains a reflex to adopt systematically, especially in mixed municipalities where houses, buildings, and agricultural plots coexist.
