Is Silicon Valley mode the Future of IT?

How cool is this? “We transform how the world builds software. Companies with the Silicon Valley state of mind are pushing the boundaries of human behavior and how business gets done. They’re nimble, software-driven, and competitive. Many are market leaders. But there’s no monopoly on good ideas. Traditional businesses have been successful for a reason. They have experience on their side. Yet they’re eager to embrace the digital future.” “Do your teams release software to production weekly, daily or every hour? Do you practice software development with tools, process, and culture that can respond to the speed of market and customer changes? Agility allows you to experiment with new business models, learn from your mistakes and identify patterns that work. In every market, speed wins.” “Cloud Native describes the patterns of high-performing organizations delivering software faster, consistently and reliably at scale. Continuous delivery, DevOps, and micro-services label the why, how and what of the cloud natives. In the most advanced expression of these concepts, they are intertwined to the point of being inseparable. Leveraging automation to improve human performance in a high trust culture, moving faster and safer with confidence and operational excellence. This is the cloud-native advantage.” Hallelujah! Everything is there that modern IT should be: Cloud, DevOps, Continuous Integration, Silicon Valley, Agility, Speed, New Business Models, Automation, Improving “Human Performance”, Operational Excellence, even Safety… You can read a lot more of this stuff here touted by Pivotal, but Pivotal is not alone in preaching this modern IT gospel. I’m just using it as an example to ask a basic question: Is the role of IT to build and deploy custom applications quickly “in Silicon Valley mode”?   Is this the future of IT? There are several examples of large companies that are carving out small “cool” entities within IT and adopt the technologies and mindset described and claiming success in conferences. So from that perspective the answer may be yes, this is the future, this is how IT should behave. There are however a number of caveats to consider:

1. Custom development has the highest TCO, regardless of development method. This is why most organizations have decided to transition to SaaS solutions and allow only configuration, no customization unless the application is unique to their core business and not available in the marketplace. Plus, do you really need continuous updates to your accounting and billing systems? For a very large subset of IT systems in the Enterprise stability may be more desirable than agility.

2. A modern view of IT organizations is as service orchestrators. In this view, IT can find partners that represent opportunities to buy services for less than IT could provide these themselves and can then capture the value in the form of cost reductions or improved capabilities. In this scenario, IT does not see itself as having unique capabilities as a source of competitive advantage. Rather it is the tapestry of deals that make it unique. Custom development can be seen as a capability that can be bought where needed for only those applications that are deemed strategic. Developing this capability in house, maintaining it and keeping it “world class” may be cool, but philosophically runs counter to this model.

3. Where I have seen this applied in IT organizations, these groups occupy a niche positioning within IT. A select cool group that does things “differently”. The intent is often to incubate these methods, practices, technologies and tools in a smaller setting, prove them out and then scale these efforts to encompass larger areas within IT. In the smaller setting many of the constraints that make enterprise IT complicated are avoided, such as compliance and security constraints, interoperability, large legacy systems etc. Can these groups really scale without causing massive disruption? And if so, is it desirable for these to scale (see other points) or should they really remain geared towards point solutions bringing competitive advantage, such as may be the case for systems of engagement?

4. Despite all the hype and all the improvements that can be made in most IT organizations with regard to their development capability, ultimately IT is not about “building software”. IT is about providing technology services of which building new software is only one. Does IT as a whole become significantly better if it addresses the shortcomings in their development capability using these methods? Perhaps in some cases that will be true, but beware that careless subscription to these trends may actually increase cost and reduce service delivery effectiveness. Case in point: consulting organizations have entire practice areas dedicated to migrating application portfolios onto these latest platforms. A lot of work/investment – but does it improve service or lower cost?