Traditional architectures and processes inhibit the flow of value and introduce blockers
Organisations transitioning from waterfall processes typically have heavyweight governance processes in place. This ensures development is in-line with technology strategy and adheres to a rigid set of detailed rules regarding design patterns, approved technologies and target architectures. This creates hierarchical decision-making structures with long lead times for approvals and removing the ability of teams to innovate. A framework of high-level guardrails and empowering teams with the accountable freedom to make decisions are key to unlocking pace. The role of the architect in this operating environment shifts considerably. Rather than being rules gatekeepers or the Big Up Front designers of the target state, they need to become the pragmatic shepherds of solutions. This can be achieved by aligning them to high level guardrails and developing just enough up front (intentional) design to avoid having to unpick big decisions later. This approach gives delivery teams the latitude to innovate at a detailed (emergent) design level in Sprint. Traditional architectures present many brakes to bringing new solutions to market. Environment provisioning, access to data held in legacy Systems of Record, and cumbersome routes to live processes can all slow down the pace of delivery. On-premises infrastructure has long been a blocker to agile delivery, with provisioning of development, test and production environments often taking months. The shift to cloud hosting and containerised solutions offers the promise of self-service automated environment provisioning to drastically cut down lead times. However, take care to ensure appropriate monitoring and controls are in place to avoid spiralling cloud costs through the exponential growth of environments. Legacy systems of record often lack high levels of availability, capacity and real-time integration required to support 24x7 digital operations. As mentioned above, these systems are not well suited to agile change and short release cycles, so often introduce long lead times. Modern architectures use data streaming technologies and microservices to liberate data from systems of record, breaking the dependency on slow moving systems and unlocking pace. Delivery into production can be hindered by throwing solutions over the fence from development into operations. A DevOps culture which brings together the development and operational teams into a collaborative delivery structure unblocks cadence and drives the upfront design of security and operational monitoring to deliver and operate solutions in a secure and resilient manner. Bringing these teams together builds the trust required to automate many of the tasks in the route-to-live and further reduces the time to market of new solutions.