Our cloud-native data platform enables companies to quickly and scalably create individual data-based applications without having to build up their own expertise. It provides direct access to unique geospatial data, machine learning routines, and analytics modules that enable even non-experts to perform custom analysis.

Our technology combines individual modules for geodata processing ("Enigma"), ML modeling ("Edison") and applications for use ("Colombo"), which we are constantly developing. This reduces your effort to build and maintain your own infrastructure while guaranteeing full data protection and control over the results: Your data remains your data, your model is guaranteed to be your model. You benefit from continuous updates of the base data and growing functionality and get access to unique expertise when needed.

Data and models are documented and versioned. Data is included in four levels of detail (LOD), depending on their pre-processing. Processing includes advanced methods and can handle multiple GIS formats as well as EO data. LOD 3, as the highest level of complexity, includes predictive analytics with ML and AI, such as price level prediction.

We have developed our application to simplify spatial data processing for various use cases in the real estate industry.

Our cloud-based application framework colombo is an infrastructure of multiple microservices and applications designed to provide new spatial insights for location decisions - hence we named the framework "colombo". It gives you the flexibility to store individual spatial strategies, integrate individually hosted data interfaces and - last but not least - it is a lot of fun to work with. The infrastructure is designed as a modular framework to support the development of industry specific modules. Currently, we offer a web application with three modules oriented to specific processes in the decision-making process: Scout, Manager and Profiler.

(Cristoforo Colombo: Italian explorer 1451-1506)

Edison is our internal data science platform and provides an environment where machine learning models can be created, validated, compared, stored in a documented manner, and deployed in a guided, semi-automated fashion. This allows our data scientists to minimize the effort required for ML analysis and its maintenance, and provide you with customized solutions quickly and efficiently:

• Preset modules for efficient training of ML models.
• Automated reporting and spatial mapping
• Provisioning for API
• Deployment in the application



Every day, we deal with data in different formats in different places around the world. We bundle the large number of methods to apply for processing and refinement into an ETL framework that we jokingly call our "spatial coding engine" Enigma. This allows us to perform an initial evaluation for any location in the world at the push of a button and is used for e.g. network analysis, raster analysis and for the evaluation of satellite images. (Enigma: mechanical encryption machine 1918)

All data is stored and managed by us in the cloud to ensure 99.2% availability of your applications. We integrate individual databases per organization on request, so that your data can remain stored locally if required. We secure this data treasure with a bulwark of measures to protect our, but also your data against external access. 

We have extensive experience in working with and processing geodata. With our application we want to make your work easier, therefore we also support the seamless integration of new data sources or internal company data. Your organization's data is stored in its own database and will not be seen by other users - unless you explicitly request it, e.g. for marketing purposes. We currently offer more than 200 layers with 30 million data points in a resolution of 25 m/point for Switzerland. According to the complexity of the data processing we distinguish so-called "Level of Details" (LOD).

LOD 0: Raw data with low information content
LOD 1: Processing of a single data set (derivation)
LOD 2: Combined processing of different data (merge)
LOD 3: Synthetic or modeled information (our own proprietary models)

Just a few examples of the uniqueness of our datasets include:

• Accessibility index
• Rent index
• Sunshine index
• Building and neighborhood typologies

The analysis of the urban structure, the "urban morphology", occupies a special status with us. With our morphological analysis we are able to identify and quantify structural features of the built environment. This method was developed by us at ETH and licensed by us. The analysis can be implemented for most regions of the world and subsequently used for predictive procedures in "Edison", e.g. it has been used in Indonesia to predict real estate prices.