The Greater London Authority (GLA) manages a range of crime and community safety datasets from key agencies in a secure database.

Groundwork GDS has been commissioned by GLA to redevelop the SafeStats site as a more stable, transparent and future-proof system with user need and efficiency at its core. The database centrally hosts 30 million crime and safety records in London from 2015 onwards, contributed to directly by nine participating agencies (MPS, BTP, LAS, LFB, ED, TFL, RINL etc). Over 300,000 records are added to the database each month, accessed by over 300 users from these agencies and London Boroughs.

For latest updates see the Safestats blog.

Key features of SafeStats

• Fully open-source technology with data security at the heart

• Scalable architecture hosted in Amazon Cloud with built-in disaster recovery and access controls

• Unlimited user logins, secure onboarding and access control via OAuth2 protocol

• Profiles and user management portal: users are associated with organisations upon sign-up. Administrators can be designated for each organisation, who can then control and manage users within their group. Users can also share their details if desired and explore their own and other users' saved queries

• User access to the system and datasets controlled by profiles: Super Admin has the ability to restrict access to specific datasets and fields, including spatial detail (e.g. hide geographical coordinates)

• Flexibility to add/amend datasets: Super Admin has full control of the data through the frontend. New datasets can be added to the system and made available to the users through profiles. Lookup tables can be associated with a dataset by defining the join through the frontend. The interface also allows users to hide fields from the query interface and/or the output
  
  
  
• Data Explorer, Query Builder and Map/Chart View for data analysis and reporting
  
  
  
• Customised queries: data can be explored and queried by the user based on categorical, temporal and spatial filters on single or multiple datasets at once. Queries can return data grouped or ungrouped by multiple temporal and spatial criteria. The user has the ability to save the query, join additional tables (e.g. population projections) and download the data
  
  
• Spatial filtering: based on their profile restrictions, the user can select custom spatial filters, using a mapping tool, which allows hierarchical selections through an intuitive interface
  
  
• Configurable data catalogue driving geospatial analysis and visualisation charts: users can visualise queried data directly on a map interface. Data grouped by spatial geographies (e.g. Local Authority, LSOA, etc.) can be dynamically filtered and explored both on the map and in charts. The user also has a control on certain aspects of the graphical output to export custom reports directly from the application. A proximity analysis tool allows users to find a specific location and query the database to find crime levels within a specific radius of the searched location