Data Architecture principles
3. Data is shared
- Last reviewed
- 4 June 2026
- Owner
- Head of Architecture
All services have data that is useful to users outside the service boundary.
Why data is shared
Maintaining data in a single service then sharing in response to business/service needs has several large advantages:
- the speeding up of data collection, creation, transfer and use
- the improvement in quality and efficiency of decision-making based on timely and accurate data
- the reduction in costs of data management by eliminating duplication and synchronisation costs
- better user journeys through re-using data DfE already holds rather than asking users to re-key
- faster service development through re-use
How data is shared
This principle is related to Data is an asset, and Data is obtainable, so all teams should understand the relationship between the value of data, sharing of data and accessibility to data.
Our digital services and legacy applications must conceptually comprise a single ‘shared environment’ across which data is easily consumed between services.
We must adhere to the common set of policies, procedures and standards for data management, discovery and access agreed by data owners.
Services must design appropriate interfaces to their data to allow other services and users to consume their data.
Services should assume that DfE will need to analyse data created/collected by that service and should use the common patterns developed.
Data should be consistently defined throughout DfE, using definitions that are understandable and available to all users. A central data dictionary should be created and maintained.
For data to be discoverable, we must maintain a searchable central metadata repository, including data elements, data models and other metadata.
Where legally permitted, data will be available externally to further the business of government and to support wider citizen and business use of DfE data.