Yesterday's rigid, siloed IT organizational structures can no longer meet the imperative to rapidly deliver high quality software.
If your IT organization could take a ‘selfie,’ how would it look? Would it illustrate a unified dynamic team driving digital transformation—a picture of IT functions coordinated into a seamless process delivering continuous value to customers? Or would it paint a more somber picture, a department locked in time just managing the status quo?
Personally, I think the majority of enterprise IT selfies are similar. Like those class pictures from grade school we keep in the attic, they’d illustrate neat rows of IT groups all doing great jobs within their own technology domain. It’s looked and worked this way for years – but there’s a problem. The organizational picture as a whole is flawed, which, in a business world addicted to software, is fatal.
The IT organizational-self is flawed because rigid structures can no longer meet the imperative to rapidly deliver high quality software. That’s why leading enterprises are turning to DevOps, which, with its focus on cross-functional collaboration, is seen as the organizational change stimulus needed for businesses to thrive.
But beyond the DevOps hype, where do organizations start the journey of IT transformation? As with anything ‘culture-centric,’ it’s about modifying sets of behaviors via strategies, which, at first glance, may appear counterintuitive.
|“When rigid structures don’t facilitate collaboration, any hope of delivering high-quality digital experiences to customers across multiple channels is just a pipe dream.”
Re-engineer Your IT Organizational Structure
Back in 1996 a programmer named Melvin Conway came up with the rather astute observation that any organization designing systems and applications will be constrained to produce systems that reflect the communication structures of the organization. Think about it. IT organizations are now being called upon to deliver highly composite and integrated API’s and application ecosystems, yet the teams and departments supporting them often don’t come close to reflecting the architecture of the applications. Worse still, when rigid structures don’t facilitate collaboration, any hope of delivering high-quality digital experiences to customers across multiple channels (each supported by a myriad of technologies) is just a pipe dream.
Many leading organizations have caught on to this and, in response, are initiating what I call a reverse Conway maneuver. That’s not a new wrestling hold, but simply a realignment of team structures according to their digital transformation strategies – usually, and not surprisingly, around mobile engagement and APIs. But the maneuver doesn’t end here. Digital leaders continuously review value-chains (a.k.a. the software development lifecycle) for any people and process bottlenecks that impede the flow of value to customers.
Rethink Your Technology Talent Pool
In the past, the hiring approach at most companies has been to follow the technology, recruiting or reskilling individuals according to the latest technology trend and wizardry. This is all well and good, but the tech landscape (especially in development) is like a hot and cold all-you-can-eat buffet. What’s cool and in-vogue today maybe be passé tomorrow. Therefore, seek out skills that are transferable across the organization, realizing the test of a true ‘team player’ won’t necessarily be found in impressive looking LinkedIn profile – it’ll be illustrated in how they showcase expertise. So for software engineering, seek out the guns who have proudly contributed code to an open source project, participated in a hackathon event, or even tried (and failed) in the startup game.
Having tweaked the structure, and the hiring policies to support it, all should be good, right? Well, not necessarily. Good talent won’t stick around long enough to deliver any benefit if bad behaviors and crazy incentives linger. Or if they do, they’ll pick up bad habits themselves. I continue to see many examples of this, with bad practices such as continuously rewarding teams for stopping software releases, or giving bonuses to the team creating the most application widgets. (Yes, Dumb and Dumber continues to play at an IT movie theater near you).
Confront Your Technical Debt
Many enterprise IT organizations continue to be plagued by technical debt – the price businesses pay for poorly designed and defect-laden software. So, before you take yet another technology toolset leap-of-faith, my advice is to carefully look at your debt position using people, process and technology dimensions. If, for example, the great development team you’ve hired and nurtured is still being pinged on both speed and quality issues, perhaps it’s time to confront the physical infrastructure constraints and contentions that inhibit reliable testing, and the process bottlenecks holding up releases.
Of course as with anything common sense, all these DevOps-style approaches can be ignored. Perhaps all you need to do is gloss over shortcomings with team reshuffles, or form grandiose sounding workforce transformation projects. But will these really be effective, or are you just papering over the IT cultural cracks to present a much rosier picture to your business peers?
Great CIOs and modern digital leaders never ‘photoshop’ the picture. Whether we call it DevOps or just high performance IT, they lead from the front –demonstrating new sets of behaviors themselves. Those who don’t and leave the organizational selfie taking too someone else run the risk of becoming irrelevant business caricatures.