For many businesses, the traditional IT development lifecycle and transition to production has historically been one of waterfall design and testing processes, followed by resilience and failover testing, before delivery to live. Stability is the key word; there should be no impact to live services.
The new world of cloud, social media and mobile is one of fast-paced change, with short development cycles, iterative releases, and incremental functionality, often in response to customer feedback. Many challengers and innovators in this area are used to operating at a pace that suggests limited exposure to delivering in a highly regulated environment, or in dealing with data protection, assurance issues, or security. The demands of this approach and the sheer pace of the propagation of change through an organisation (and its test environments) can be a culture shock for organisations unused to iterative and agile development.
Trying to ensure that you stay involved with the latest digital developments, while also continuing to support the traditional, highly critical, IT systems can prove challenging. There are however three things that might help.
Make it easy to try new things. Most CIO’s would share the ambition that the IT department should be the easiest route to create a proof of concept. Today, where all you need to get access to Cloud computing resources (like Amazon Web Services) is a credit card, this is not easy. However, it is possible to quickly set up a proof-of-concept within a virtual environment – with a fixed-duration, naturally. There are many “proofs of concept” that can be tested when your organisation wants to try something new.
Be adaptable and innovative
IT solution design and delivery, is inherently a creative problem solving activity, housed within a structured approach. You will no doubt have experiences of programmes where the ultimate successful delivery hinged on a new way to do something that nobody had tried before. However, you will also have built up a wealth of “near misses” and a healthy caution towards the cavalier. We need to move away from “that won’t work because …” and “we tried that before and it didn’t work” to a mentality that is more: “How can we make that work in a way that is supportable?”
But be rigorous
To be responsive to the needs of the business requires robust (but slick) internal processes in the IT department. Standard images, virtual builds, and automation. It can’t take three weeks to give some developers a POC environment when they can get one from a cloud service provider in a few hours. And the more involved you are at the start, the more easily you can ensure that the discipline of activities like release processes between environments, test data integrity, and security, are adhered to, instilling the rigour that you require later when transitioning into live.
In summary, be ready and prepared for the challenge to be more agile and responsive. And of course, it is worth repeating that a clear set of requirements and measureable success criteria from the business – not to be confused with a desired solution – will avoid an unnecessary conflict during the design and implementation when certain requirements are competing for priority.
David LukeBack to articles