Government: Delivering a post-Brexit reusable test automation framework

The background

When goods are imported into the EU, traders must declare them prior to arrival at their port of entry so that their security risk profiles can be determined and assessed. The Import Control System (ICS) is a Europe-wide platform to enable the online processing of safety and security declarations for these goods and automates procedures so traders only have to submit information about goods only once.

Built on legacy technology, it was projected that the UK’s ICS would not be able to process the anticipated increased volume of imported goods (89-90 million declarations and submissions per year) post-Brexit.

A new system was required and our government client initiated the build of the greenfield safety and security system to bring together all elements of the goods imports and declarations. This system had a fixed deadline and was scheduled to go live at midnight on the 31 December 2020.

The solution

Understanding the requirements and culture fit

Tecknuovo was engaged to provide an independent testing capability for our client’s Customer and International Trade (C&IT) Delivery Group partnering with the incumbent testing supplier. We rapidly deployed a team of multi-skilled testing specialists to lead and support on test management, component testing, performance testing, SIT, OAT, and end-to-end testing. Our flexible engagement model enabled us to ramp up/down the team at short notice based on the needs of the project and without disruption to the tight timelines.

From the start we actively engaged with inhouse and third-party teams to understand the technical and business requirements and ways of working, so we could define the overall testing strategy as well as the best methods for communication and collaboration. As a team, we adapted our approach to suit the multi-supplier ecosystem and the clients mix of Waterfall methodology and Agile delivery aspirations.

The Tecknuovo team introduced many new Agile best practices, including ceremonies such as the Three Amigos, bringing together leaders from the test team, dev team and BAs to ensure a consistent understanding of the required functionality and a common focus on the desired outcome. This is a key component of our One Team approach.

Overcoming unexpected obstacles

The team identified the need to set up a mock API in a headless Lambda server. No one within the client site had prior experience of doing this, so our team took up the challenge using our DevOps and technical knowledge. Within three weeks – and without impacting project timelines – the Tecknuovo team developed and delivered a mock ReadyAPI, deployed and applied in the headless server instance within the AWS environment.

We fully documented the entire process and created ReadyAPI Mock standards making it available to the entire programme via Confluence so that the knowledge and learnings could be used for future projects across the organisation.

Security against system failure

Due to the increased volume of anticipated declarations – over three times more than were run though the ICS annually – additional resilience needed to be built in to protect against system failure. During the Operations Acceptance Testing (OAT) stage, Tecknuovo built a secondary system. Working in a “Source/Replica” relationship, this second system was designed to pick
up any traffic from the declaration processes in the event of a network outage, server failure, or an unresponsive service.

Early Non-Functional Testing (NFT)

Traditionally performance and OAT testing takes place near the end of a project. This leaves the client open to project delays – if a defect is identified close to the go-live date, it could push back the whole project.

To mitigate this risk and identify issues as early as possible, our team introduced a light version of performance testing early in the project, at the component stage. This identified a defect that took the technology vendor three and a half months to fully resolve. However, thanks to our early NFT, there was sufficient time for the vendor to issue a hot fix which
was applied and so there was no delay to the project.

Building in reusability

We created a brand-new test automation framework developed specifically to ensure reusability across all of the organisation’s projects. With minimal changes, it can be applied to deliver any API-based testing across the client’s portfolio of work.

This is a huge value add for the client as it reduces the work required in future builds enabling new APIs to be produced 50% quicker on average, saving both time and money.

Upskilling as default

In order to ensure a seamless transition from Tecknuovo inhouse staff, all our teams provide continual knowledge transfer throughout our engagements. When it came to the final handover on safety and security, our Test Manager curated a plan of knowledge transfer sessions that covered all aspects of the new API Automation Testing Framework. Any learnings the team made during the project were also documented in full and uploaded to our client’s collaboration tools as standard.

Our results

Through our Agile ways of working and One Team mentality, the Tecknuovo team were able to deliver story points at a 25% increased velocity per team member against the baseline performance.

In addition to the work delivered by the Tecknuovo team, our continuous knowledge transfer and clear documentation meant we were able to leave the client with Zero Dependency on us once the engagement completed. We also ensured future savings of time and money for the client, thanks to the reusable and robust new API automation testing framework we created.