4. Build the service using the agile, iterative and user-centred methods set out in the manual
The team works according to agile methods without being slavish to a particular methodology. We tend to talk about small ‘a’ agile rather than big ‘A’ Agile. There are daily stand-ups, sprint planning sessions, retrospectives (every two sprints), a classic Kanban wall in a public space and project artefacts on every spare bit of wall possible. We split our activities into epics, user stories and tasks – with tasks tending to be the actual technical activity required to deliver the functionality needed for a story. Every epic & story can be traced back to user needs identified either during discovery or during user research during the Alpha or the Beta.
There are open show & tell / demo sessions at the end of each sprint which Programme Board members are encouraged to attend and if you are reading this you know we blog about pretty much everything.
The team is co-located so there is a steady underlying murmur of communication that is enhanced by liberal use of Slack channels (including short week notes from everybody). As for examples of learnings from user research during the Alpha the addition of ‘spark-lines’ to the headline figures was based on feedback that the numbers were not helpful unless they were put in wider context.
The major immediate change in the UX/UI of the Beta regards the taxonomy landing pages (i.e. Inflation and Prices) where there ‘hero’ space with links to key indicator figures/time series and separately to the most reason Statistical Bulletin did not test well so it has been re worked to make things significantly clearer (this is an element that will again be tested again as soon as possible).