Approximately 80% of data has a location attribute associated with it – and that location data provides a connection with the physical world.

For Generali Real Estate*, the addition of alternative forms of data, such as spatial data, created greater context for its data and helped to power highly accurate AI-driven insights for data-driven decision-making.

Let’s take a closer look at their journey.

Disrupting Traditional Real Estate Investment Decision-Making

Generali Real Estate is one of the world’s leading real estate asset managers. Headquartered in Italy and with operations across Europe, the company has €36.9 billion assets under management (Q2 2024). When Generali Real Estate became one of the first real estate asset managers to establish a dedicated division for AI and machine learning (ML) innovation, its first task was to disrupt the traditional decision-making processes that usually inform investment strategies.

For example, standard real estate metrics often don’t reveal the reason for significant variances in the value of assets, even when similar assets are only within a few streets of each other. The team discovered that as much as a 60% change in value, observed over seven years, could not be explained using classic real estate metrics such as prime rent or capital value.

To address these challenges, Generali Real Estate developed City Forward^®, an innovative cloud-based location intelligence platform that helps real estate professionals and others make smarter decisions powered by highly accurate AI-driven insights.

“We wanted to use alternative forms of data, especially spatial data, to address these problems,” says Costanza Balboni Cestelli, Head of Data Intelligence & Innovation for Generali Real Estate. “Ultimately, without data context, there is no such thing as AI in the field of location intelligence.”

Data scientists for City Forward needed to feed the ML models, but they faced challenges, including:

Standardizing data coming from different data sources
Verifying the accuracy of the data
Feeding data to ML models with maximum accuracy and consistency
Enriching in-house data with accurate third-party data to feed models and provide lift

Powering Success with Location Intelligence and Data Enrichment

In their search for the right data partners, the City Forward team wanted to balance global and local data sources and ensure consistency, quality, and scalability. They selected Precisely for their extensive expertise and heritage in location intelligence and data enrichment.

“We started using Points of Interest data from Precisely, along with other types of data such as social demographic, satellite or real estate data, that can also be transaction based,” says Balboni Cestelli. “Then we paired them to understand what effect those variables have on market attractiveness or the value of an asset. Precisely provides us with access to accurate, consistent, and contextual enrichment data that helps power our AI/ML models in a way that is both scalable and reliable. We started with 20 variables and immediately, point of interest (POI) and proximity stood out as one of the most significant. Cities and geographies influence, and are influenced by, everything that goes on around them. The idea was to build a structured and scalable database that contains point of interest data.”

The City Forward platform leverages the Precisely portfolio of market-leading geo addressing solutions, alongside data from other third-party providers, to deliver comprehensive information on business locations, leisure hot spots, and other geographic features – revealing hyper-local insights on real estate assets and more. Because Precisely assigns a PreciselyID to every address it geocodes, it’s easy for clients to analyze data for attributes that relate to specific locations.

For Generali Real Estate, City Forward paved the way for a quantum leap forward. “This solution led to a new level of precision in the real estate industry, and we pioneered the use of alternative data for real estate. We are testing more use cases from retail to urban planning, to ad industries, all over Europe,” Balboni Cestelli says.

Today, City Forward is Europe’s largest, most varied, and most granular data infrastructure. The application uses more than 800 variables and more than 30 ML models, bringing unprecedented granularity to forecasts.

The Future is Bright – and Scalable

Beyond real estate, City Forward is scalable across industries and geographies where it’s used in more than a dozen use cases to shed light on sociodemographic information, consumer habits, web data, ESG (environmental, social, and governance) reporting such as CO2 emissions, green areas, criminality, points of interest and territorial data, people mobility, traffic and tourism flows, and satellite data.

“We have been working with Precisely since day one, and we still do, and we will continue to. Our 30 ML models have an average accuracy of 95%. Our satellite modules are extremely useful as they have an average accuracy of 80%,” Balboni Cestelli says. “More than 400 colleagues are using City Forward data for real estate operations, as well as a few other clients in retail and the public sector. Going forward, we’re exploring how to incorporate GenAI, use computer vision or other technology to enhance images, and increase the number of use cases where location intelligence can change the game.”

* In recognition of its innovative work, Generali Real Estate received the Data Integrity Award for Best AI Impact. For more information about Precisely, please visit www.precisely.com.

The post Without Data Context, There is No AI appeared first on BigDATAwire.