Update 10/31. Attending multiple conferences this fall convinced me that there is a broad based movement towards product thinking among I&O teams. The term “platform team” is increasingly heard, meaning an infrastructure shared service managed with product team principles. 

As an analyst at Forrester, I talk to a lot of IT professionals. One of the most common inquiries lately goes something like this: “We’ve piloted DevOps and Agile and like the results. However . . . ” and then come the “what-abouts.” What about ITIL, the NOC, the auditors, the service desk, and perhaps the biggest what-about of all: the infrastructure and operations (I&O) organization itself. What to do with this organization that for years has run on tickets and work orders executed by specialists?

It’s a hard question. The next-generation I&O operating model is a current research program for me. Clearly, the old model is under siege: developers armed with infrastructure-as-code, product teams taking over traditional Tier 2 support, automation and AIOps eating away at Tier 1, and relentless pressure for faster delivery. Ticketing everyday work — putting it into a queue — is just not good enough. Six-month lead times are utterly unacceptable. So what are the alternatives?

Part of the answer lies in the old adage, “if you can’t beat them, join them.” The integrated, semiautonomous, multi-skilled product team is here to stay. It is the most effective means for discovering and delivering value to customers. I believe the integrated product team is the most important economic unit of the 21st century. Can the old I&O dog learn these new product team tricks?

I think the answer is yes, but that’s easy to say and harder to do. There are pitfalls. You can take a team of server engineers, or DBAs, and call them a “product team,” but what good does that do? You probably still just have a functional silo, now acting like a cargo cult. True organizational transformation is harder than that. In order to don the product team mantle, you (and your teams) need:

  • Empathy. This is the biggest difference between a functional silo and a true product team. Do you put yourselves in your customers’ (or users’ or stakeholders’) shoes?
  • UX and design thinking. Grizzled I&O pros may roll their eyes at such a “soft” concept as empathy, so let’s make it real. Have a look at the leading practices in product development: user journey mapping, persona analysis, design thinking, and so forth. How can you integrate these concepts into your I&O service offerings?
  • Full lifecycle. You discover it, you design and engineer it, you build it, you run it. These principles apply to I&O services as well as application or feature-level work. They may require interpretation; setting up guardrails and approved templates for a cloud offering is different from engineering an internal SAN service, obviously.
  • Automation and self-service. This is especially for infrastructure teams — APIs keep your customers happy. If it’s basic, entry-level resources they need (e.g., development servers), give them templated, standard environments instantly available on demand (with a proper budget code, of course). And think creatively about automating not only infrastructure provisioning but change and incident management (e.g., via ChatOps), NOC and service desk onboarding, and more. Down with forms, tickets, workflow, and manual reviews. Up with APIs!
  • Enterprise service management. So you have an ever-increasing portfolio of internal product teams, offering a wide variety of services to each other. How can people find what they need and access it, preferably self-service? This is where enterprise service management comes in. (Updated Forrester Wave™ forthcoming!)

I cover all of this and more (including suggestions for I&O “products”) in “Update Your I&O Operating Model With Product Team Principles.” Let me know how your product team journey is going!


Credit where due: I recommend Designing Delivery by my friend and fellow Minnesotan Jeff Sussna for excellent coverage of this and related topics. I also valued recent Twitter interactions with John Cutler of AmplitudeJosh Arnold of Black Swan Farming, and Kevin Brennan of Product MVP Consulting; they all have thought deeply on this topic. Damon Edwards‘ concept of “operations-as-a-service” in the recent Seeking SRE was an “aha!” moment. And I was pleased to see that Clint Boulton of CIO.com has also started covering this trend, here and here.