Department for Education
How Talk Think Do Helped the Department for Education Modernise and Adopt a Cloud Native Architecture
The Department for Education (DfE) is the UK government's department responsible for education and child protection in England. Talk Think Do helped the DfE modernise its IT systems and embrace the cloud.
The Problem
The Education and Skills Funding Agency is responsible for allocating £60bn of funding to UK schools. It also provides apprenticeships and oversees funding for academies and further education. Learners of all ages across the UK rely on the organisation to help provide opportunities and training to develop essential skills.
The Calculate Funding Service was established to provide a joined-up system to allow policy specialists to design, prototype and implement allocations. However, the existing on-premise solutions were no longer coping with the load generated.
It was clear moving to the cloud was a potential solution, but there was a lack of cloud experience within the DfE team to carry this out effectively.
The Solution
Working within a Government Digital Service (GDS) process, we provided solution architecture from discovery through to design.

We proposed an Azure serverless architecture, using Cosmos DB and Azure Functions, which would provide the ability to quickly undertake policy calculation during busy periods.
We produced a working proof of concept and then helped to build a delivery team and upskill the existing internal support team.
The Results
Funding schools and other educational organisations and initiatives is a crucial part of the work done by DfE and its Education and Skills Funding Agency. Moving to the cloud was crucial in ensuring these funding processes functioned optimally.
Talk Think Do provided value on two fronts:
- Designed and architected a cloud-based solution to modernise the Calculate Funding Service for the demands of modern policy calculation
- Upskilled the internal team ensuring the DfE had the capability to maintain and evolve the system independently