Operationalizing the shift toward cloud native provides a new, enormous complexity to organizations. Yet, those challenges are more about changing people and culture than technology and capabilities. To succeed in becoming cloud native, organizations need to define business objectives and apply technology to achieve them. Operations enable the delivery of those objectives. To succeed with their transformations, organizations must blend the constant curiosity of an engineering mindset with the collaboration of DevOps culture. To help cut through the marketing noise, HFS will discuss the Seven Pillars of Cloud Native Operations.
To help more organizations finally capture value from their cloud investments, we need to drive the discussions back to business outcomes. Providing more clarity on the operational objectives is a significant part of this. Therefore, we have identified seven pillars of cloud native OneOffice operations in Exhibit 1. We don’t consider these topics a linear progression, but we aligned them to reflect the transformation life cycle. These pillars provide a refinement of HFS’ OneOffice narratives. The very nature of OneOffice is to overcome organizational boundaries and silos. This goes beyond bringing the front, middle, and back offices closer together; it is in equal measure about finally aligning IT and business operations.
Source: HFS Research, 2022
Achieving data-driven outcomes is at the heart of the HFS OneOffice concept. Organizations increasingly seek experience-led outcomes ranging from customer to employee to ecosystem experiences. To achieve those outcomes, they must align their data needs to deliver on business strategy. Take the energy major changing part of its business model from fossil fuels to decarbonization. For enterprises to succeed with this transformation, data and application modernization must run in parallel.
The biggest challenges are to leverage vast data assets spread across diverse geographies with an unparalleled velocity. Without cloud native technologies, such a transformation journey was simply inconceivable within the timeframes available and at an acceptable price. Organizations need help envisioning business outcomes, such as operating model transformation.
Cloud native demands a major step-change in ways of working and requires an agile, market-speed model. This includes DevSecOps practices that can scale on an industrial level, a shift to a product mindset, agile delivery, and a culture of innovation and continuous learning. Yet, there is no such thing as the cloud native operating model template. Those models must be tied to business outcomes, and these are specific to every organization. However, when tying technology objectives to those models, organizations can leverage a broad array of composable building blocks. Access to composable capabilities is crucial, as there is no end-state for cloud native transformation: We need to acknowledge that transformation is a permanent process.
Given the propensity of the marketing of the supply side to focus on technology capabilities, it might sound counterintuitive that the journey toward cloud native is shifting the emphasis to the business side rather than the IT side. To get to the envisaged business outcomes, organizations increasingly look to industry clouds—industry-specific, modular, and pre-configured solutions that serve as composable building blocks. Both the hyperscalers and major service providers are building out those solutions. Yet, implementing industry clouds requires a new operating model. Thus, without envisioning new models, cloud technology will remain a horizontal and largely infrastructure-centric play. To capture value, enterprise leaders must align technology transformation, such as industry clouds, with business objectives. Without that linkage, investments will be largely wasted.
We see among the service providers that this is not leading to a convergence of IT and business operations. Rather, they are bringing the capabilities and toolsets closer together so that centralized teams can design solutions and thus overcome some of the operational silos. Within the primacy of business solutions, operations are shifting toward persona-based solutions. That means having the capabilities to better support specific stakeholders such as product teams, site reliability engineering (SRE), and DevOps engineers. Those newly evolving personas and teams will drive business-centric solution design, but they won’t drive converged operational platforms.
One of the biggest cultural changes of the cloud native paradigm is the shift from a project to a product mindset. The days of plan–build–run of multi-year transformation projects are numbered as cloud transformation forces IT leaders to act fast, think outside the norm, and embrace the blurring boundaries between digital and physical worlds. Horizontal technology teams that focus almost exclusively on optimizing costs and providing resiliency are giving way to product-oriented teams focusing on driving responsiveness and agility. The central element of this new mindset lies in aligning IT operations with an entire product life cycle rather than being tied to a project’s start and end date. Thus, IT and, ultimately, OneOffice operations will align with business objectives.
Therefore, we need much more clarity on the business objectives and the operating model transformation that must underpin cloud native operations.
Inherent in this new product mindset is the non-linearity of workflows and the interdependency of information. This is also what makes for complexity in cloud native operations. To handle these, organizations need new roles and responsibilities for their operations teams to build the skills and hire talent to deal with these shifting requirements. Top of mind is SRE. In the words of Google’s seminal book on the topic, “SRE is what you get when you treat operations as if it’s a software problem.” Everything-as-a-code concepts are central to dealing with the non-linearity and velocity of cloud native capabilities. Everything-as-code is an approach to IT operations that uses code to define and manage resources of all types. In essence, these approaches use policy files to govern how software is built, deployed, and monitored.
Furthermore, SREs build much of the automation for DevOps workflows, which is critical for linking technology to business objectives. Against this background, executives at Accenture crystallized the essence of being cloud native: “It is all about engineering, but it has to be grounded in business objectives.” But enterprise leaders must be clear in sequencing their transformation journey. It has to begin (or be the starting point for a re-evaluation) with the definition of business objectives that should culminate in designing a new cloud target operating model. Only then can technology transformation be aligned to those efforts.
To manage this new complexity, the ambition is to have a golden pane with operational data allowing operational teams to manage and understand these new cloud native solutions and their interdependencies end-to-end. The goal is to have the necessary data that allows performance monitoring and corrective action where bottlenecks or outages occur. Yet, the challenge is that this data, often telemetry data, is tied to specific domains. Therefore, organizations need data modernization in parallel to application modernizations when they progress toward cloud native. Crucially, data modernization cannot just be about customer data; it must be about all data inputs, especially operational telemetry data.
Digital trust is paramount in managing those cloud environments. Yet, being compliant does not necessarily equate to being secure. Enterprises must look beyond the compliance agenda and deploy holistic security policies to act as a foundation for a robust cloud security framework. Therefore, the ambition for organizations must be to support the DevSecOps life cycle with broad AIOps and observability capabilities. This could allow development teams to build feedback loops into their applications in just a few clicks. It is about progressing from insights to actions and, ultimately, automation. Thus, cloud native operations must go beyond basic automation and progress toward orchestration of customer experience and business outcomes. Those business outcomes increasingly focus on supporting specific personas such as DevOps and SRE engineers or business analysts. A compelling example of such an approach is Infosys’ Live Enterprise Suite for Intelligent Operations. It is compelling because Infosys has made significant progress in transforming data inputs for use in multiple domains, allowing organizations to map key performance indicators (KPIs) to processes and track the delivery of the envisaged outcomes.
The new complexity of operating in a cloud native environment means organizations should accelerate their automation and AI capabilities. For TCS, which has championed the machine-first mindset, this means allowing technology the first right of refusal to sense, understand, decide, and act on delivery issues and any problems during customer journeys. A broad set of automation and AI capabilities is providing Level 0 support. That denotes solving incidents before they reach the service desk through a combination of self-service and self-healing. On the latter, however, we must be clear that apart from managing some specific network issues, there is no such thing as self-healing as in a bot or machine fixing issues autonomously. Self-healing or notions of the “autonomous enterprise” are marketing-led descriptions of the ultimate ambition of operations. Thus, Level 0 is about proactively identifying issues and guided resolution, including automation technologies such as Runbooks and robotic process automation (RPA).
Invariably, automation can play a key role in such a machine-first mindset. Leaders should look beyond hyped topics such as RPA. We need to recall our experiences with innovative approaches such as infrastructure-as-code for provisioning, DevOps automation, or the evolution from good old workload automation into service orchestration and automation platforms (SOAP). Enterprise leaders must progress their automation strategies to an enterprise-wide view; progress toward cloud native necessitates taking an end-to-end view of processes and, therefore, automation.
As with the product mindset, innovation in software development around the concepts of SRE and DevOps is driving a fundamental cultural shift. These concepts drive software development at an industrial scale and transforming how operations work and collaborate. The aspirations of continuous integration and continuous delivery (CI/CD) break with the linearity of workflows. Evermore autonomous development and delivery teams are increasingly given continuous rather than fixed or linear goals to reflect the complexity of cloud native operations. Unsurprisingly, the concepts of everything-as-code and high degrees of automation are central to the progress of that transformation.
Yet, where we need many more insights and best practices is where those SRE and DevOps teams intersect with business operations. As highlighted in the discussion on business-centric solutions, the journey toward cloud native is shifting the emphasis on delivery and operations to a business context rather than an IT context. But how does the funky and innovative culture of SREs and DevOps work with the often-stuffy world of business analysts? The business operations side remains much more linear as workflows follow standard operating procedures, and truly cognitive capabilities are few and far between. However, not least for running processes containerized, we need a fusion of the cultures of IT and business operations. And more fundamentally, IT and business operations must overcome their siloed existence to progress toward the OneOffice.
To get to the envisaged outcomes, organizations expect the velocity the cloud offers and want to see a certainty of outcomes. Yet, cloud native means there is an inherent shared responsibility across a broad set of stakeholders to guarantee governance, risk, and compliance. That responsibility ranges from cloud providers to new operational teams and customer organizations undergoing transformation change. Thus, it is crucial to identify the right objectives and tie them to the business outcomes the operating model should deliver. To succeed in this state of fluidity, the building blocks of business assurance and operational checks and balances must be consumable—just like the cloud itself. Executives at Cognizant tempered their hope for the certainty of outcomes. For them, business assurance is predominantly dependent on governance, as there literally is no certainty. Therefore, governance must be consumable, and it must safeguard access to the scarce pool of talent that can manage those cloud native operations.
From an operational point of view, we see providers moving from service level agreements (SLAs) to business level agreements (BLAs). The intent is to work from BLAs backward to operational tasks and tools. Knowledge graphs are the glue behind the integration of all those disparate inputs and tools. Such knowledge graphs represent a network of real-world entities—objects, events, situations, or concepts—and illustrate the relationship between them. In the end, it is the relationships of all those expansive solutions and toolsets that allow organizations to progress beyond the operational silos. But it is also where quality assurance could find new strategic relevance. For years we have heard about business assurance from the quality assurance community. But for quality assurance to become a boardroom issue, helping organizations tackle this new complexity and providing those certainties of outcomes could be the emancipation the community craves. Thus, enterprise leaders should look at business assurance in the same way they tie technology transformation to business objectives: Focus on outcomes, not capabilities.
Becoming cloud native is no longer just about migrating workloads to the cloud and leveraging cloud platforms such as Salesforce, ServiceNow, or AWS. It is about a fundamentally different way of working and collaborating with data and AI as the centers of gravity. And it is about operationalizing the OneOffice mindset. To capture the value of their investments in cloud native technologies, organizations must redefine their operational strategies and place designing cloud native target operating models top of mind.
Thus, to progress on their transformational journey, organizations need not only to drive the discussions back to outcomes, they also must elevate operations as a boardroom priority. The cultural change underpinning cloud native operations is inextricably linked to talent transformation. Therefore, we must cut through the market noise on technology and capabilities. To do that we need to learn more from organizations that have successfully begun transforming their operations. With that we are going full circle on our engagements around the OneOffice and, ultimately, the OneEcosystem mindset.
Register now for immediate access of HFS' research, data and forward looking trends.
Get Started
If you don't have an account, Register here |
Register now for immediate access of HFS' research, data and forward looking trends.
Get Started
