From a Simple Question To a Data-Driven Insight

The first answers from Quanthome AI already gave a clear picture: there are tens of thousands of dwellings in Lausanne, and a significant part of them are in PPE. Some neighbourhoods show a very high concentration of owner-occupied or condominium housing, while others are still dominated by rental buildings.

Very quickly, the conversation moved from a general overview of Lausanne to a more detailed view by postal code. I asked Quanthome AI to look at specific NPAs such as 1003, 1004, 1005, 1006, 1007, 1010, 1012 and 1018, and to compare both the number of PPE units and the average price per square metre.

Quanthome AI then highlighted areas where PPE is particularly important. For example, certain zones combine a high share of condominiums with above-average prices per square metre, making them key areas for understanding demand, affordability and investment potential. Other central zones remain mostly institutional or rental, even if prices are high.

In just a few exchanges, I obtained something that usually takes time and coordination between different teams: a clear, data-backed story of how ownership and prices are distributed across a city.

Asking Smarter Questions, Getting Smarter Answers

The most interesting part of the conversation came when I started asking the kind of questions analysts ask every day: is there a link between the share of PPE in a neighbourhood and the level of prices? Do areas with more owner-occupied housing tend to be more expensive? Or is it more about location, prestige and history?

Instead of giving a vague or generic opinion, Quanthome AI did what a good analyst would do: it went back to the numbers. It built, for each postal code, a small dataset with two columns: the percentage of dwellings in PPE and the average price per square metre. On top of that, it measured how strongly these two variables move together, using a simple correlation analysis. In practical terms, it looked at whether postal codes with a higher share of PPE also tend to show higher prices, and to what extent this pattern is strong or weak rather than just random noise.

The conclusion was nuanced and realistic. There is a positive link: areas with more PPE tend to be more expensive on average. But this is not the only factor. Location within the city, views, access to services, the age and quality of buildings and the history of each neighbourhood all play a major role. Quanthome AI was able to explain this in simple language, without jargon, while still relying on real data.

In Lausanne, from a more human and opinionated perspective, this pattern also makes intuitive sense. Developers tend to build condominiums where prices are already high, because selling units one by one is more lucrative when square metre values are strong. Once these projects are delivered, they naturally attract households who can afford to buy rather than rent. Over time, this creates a feedback loop: more condos bring in wealthier residents, the area upgrades, cafés and services follow, and prices rise further. In other words, higher prices encourage more PPE, and more PPE can, in turn, contribute to gentrification and even higher prices. And this is not just a Lausanne story: at the national level, similar mechanisms are at work in Switzerland’s main urban centres, where higher price levels often go hand in hand with a larger share of condominiums and gradual gentrification. With Quanthome AI, this type of analysis can be extended from a single city to the whole country in a consistent, data-driven way.

An AI That Knows Its Data and Accepts Being Challenged

This conversation also shows something essential about how Quanthome AI works. The system is directly connected to the Quanthome data platform, which aggregates Swiss real estate data at building and vehicle level. When it answers, it does not guess; it queries structured datasets: building attributes, tenure status, transactions, prices and more.

Just as important, Quanthome AI can be challenged. At one point, I signalled that a previous answer was wrong. Instead of defending itself, the AI reviewed the filters, checked its calculations and corrected the result, explaining what had changed. For professionals, this behaviour is key: you can treat Quanthome AI like a junior analyst who is fast, tireless and able to self-correct when pushed.

What This Means for Investors and Institutions

Even though this example focuses on Lausanne, the same logic applies to any city or portfolio. An asset manager can ask which zones in their portfolio have the highest exposure to PPE, where high prices and high ownership coincide, and whether their investments truly maximise value per square metre, or where new developments are changing the balance between rental and owner-occupied housing. A bank can use similar conversations to refine lending strategies by area. A fund manager can use it to prepare investment memos that combine geography, tenure and price dynamics.

The key point is that none of this requires learning a new tool. You stay in a conversation, and Quanthome AI does the heavy lifting on top of the Quanthome data platform: filtering, aggregating, checking, documenting and explaining.

Towards a New Way of Doing Real Estate Analysis

For years, finance has had tools like Bloomberg to standardise and professionalise market analysis. Real estate has often remained fragmented, with data spread across PDFs, Excel files and local databases.

Quanthome aims to change that. By combining a unified real estate data platform with an AI that you can talk to, we make it possible to move from a simple question – “Which area has the most condominiums?” – to a city-level diagnostic, and then to portfolio-level decisions.

Quanthome AI is not here to replace analysts. It is here to give them a clearer view of reality, faster, and to help them ask better questions. And sometimes, everything starts with just one line in the chat box.