Data Architecture principles
1. Data is an asset
- Last reviewed
- 4 June 2026
- Owner
- Head of Architecture
Data is like any shared asset with value to the whole department and beyond.
Why data is an asset
Data has value because it supports decision-making; in front-line delivery, in policy making and in public accountability of the education sector. Like all corporate assets, data costs money to create or procure and then to store, access and protect throughout its lifecycle. So it's important to maximise the value, with careful, proactive management, ensuring its availability, accuracy and reliability for well-understood legitimate user journeys.
Data assets are also targeted for their value to illicit interests and so need to be protected as with any corporate asset. For all of these reasons, we must have clear visibility of where each data asset is, how it adds value and the individual who owns it on the department’s behalf.
How data is an asset
This principle is related to Data is shared, and Data is obtainable, so all teams should understand the relationship between the value of data, sharing of data and accessibility to data.
All data should be registered in DfE’s Information Asset Register (opens in new tab, DfE SharePoint users only) that records the legal basis on which we capture the data, how we process and use it, who owns it and the policies and privacy notices that apply to it. This will ensure we remain visibly compliant with current legislation.
Data must only be used for legally permitted purposes, as defined by GDPR and captured in the Information Asset Register (opens in new tab, DfE SharePoint users only).
Services must work to improve the value of our shared data assets by ensuring processes are in place to maintain and improve data quality, in accordance with the Technical standards. This includes making sure data is validated by users at point of capture wherever possible, and tracking the changes to data across user journeys.
Services must refrain from diluting the value of our data assets by creating their own versions of assets – we must agree master data sources and ensure those are always used.
We must have clear accountability for our data assets, in the form of data owners.
Our data assets must be properly managed, with responsibility for this management sitting with data stewards.
Stewards must have the authority and mechanisms to manage the data for which they are responsible.
Data stewards must ensure data quality, and should have procedures and plans to correct errors in their data and improve processes that create flawed information.
Data quality should be measured and made visible to all users of the data.