February 2025 Summaries

CARTO has announced the release of Raster support in its platform, allowing users to work with raster data directly inside their cloud data lakehouse. This marks a significant milestone in spatial data analysis, providing unparalleled scalability, flexibility, and efficiency for industries such as telecommunications, insurance, and climate analysis that rely heavily on raster data. With this release, CARTO becomes the first platform to support all aspects of raster needs, including ETL, analysis, and visualization, eliminating the need for multiple external tools or systems. The new feature is designed to address the limitations of existing cloud-based processing, which often required specialist GIS and earth observation tools to transform raster data into insights. By integrating Raster support natively within the cloud data lakehouse, CARTO aims to provide users with a seamless experience for working with raster data, minimizing risks, time, and cost savings, while also paving the way for new open data standards like Parquet raster in the cloud.

Signal propagation analysis is crucial for understanding how signals travel and behave in complex wireless networks. Traditional methods often rely on dedicated tools that operate in isolation, limiting their performance and integration with other data analytics ecosystems. CARTO Workflows has introduced a new Telco Signal Propagation Models extension package to break this silo, enabling users to analyze wireless communication networks without extensive coding knowledge. This extension provides components for line-of-sight analysis and path loss estimation, complementing existing capabilities for analytics, data processing, and data enrichment in the telecommunications industry. By integrating geospatial data like population density and mobility patterns, CARTO Workflows' signal propagation extension enhances analysis with a low-code interface, making advanced telecom analysis accessible to professionals without technical expertise. The solution is built on a cloud-native architecture, ensuring fast processing and seamless scalability for projects of any size.

Automating spatial workflows in Databricks with CARTO is essential for organizations working with geospatial data, as it simplifies spatial data processing and makes it easier to extract insights and integrate them into broader data pipelines. By leveraging Databricks' built-in capabilities, CARTO enables users to design and automate geospatial workflows natively within Databricks, ensuring seamless integration into existing pipelines without the need for external schedulers or manual intervention. This approach allows organizations to run spatial analysis on a reliable and automated schedule, improving efficiency and consistency across their Databricks environment, while eliminating the need for external dependencies and providing a cloud-native approach to geospatial automation.

Property insurance is undergoing significant transformations due to severe weather events and the rise of parametric solutions, which pay out based on predefined triggers rather than actual loss assessments. However, crime data remains an often overlooked factor in property risk assessments, with FBI estimates revealing over $16 billion in annual losses due to property crimes alone. Organizations are taking a more sophisticated and data-driven approach to risk management by using advanced spatial analytics that improve property risk assessments. CARTO plays a pivotal role in this process by offering users the ability to undertake advanced spatial analytics that enable insurers and real estate managers to make more informed, proactive decisions. The company's low-code design tool, CARTO Workflows, allows users of all technical levels to implement complex analyses with ease. By using Detect Space-time Anomalies, a feature within CARTO Workflows, users can identify vacant buildings in areas experiencing anomalously high rates of violent crime, which can be highly beneficial for real estate insurance companies as vacant properties are more vulnerable to vandalism, theft, and arson. This insight enables better risk management and helps insurers mitigate potential losses, ultimately improving the accuracy of underwriting decisions and enhancing profitability.

The BigQuery ML Extension Package is a new tool that allows users to extend the functionality of CARTO Workflows by creating packages of custom components through SQL stored procedures. This package enables seamless integration of machine learning models into automated pipelines, empowering businesses to unlock valuable insights and automate data-driven decisions. The extension package can be used to train predictive models such as customer churn prediction or sales forecasting using data stored in Google BigQuery. Users can either train their own models or import custom models trained outside of BigQuery, and run predictions, understand the model's output, and evaluate its performance. The tool provides a low-code environment for building, training, and deploying machine learning models, and offers features such as feature importance analysis and explainability. By integrating machine learning capabilities directly from Workflows, users can simplify their data-driven decision-making processes and drive better outcomes.