DevOps principles are built on an effective, continuous cycle of improvements—key factors that help drive digital transformation for companies large and small. Inserting this agile mindset into culture and operations, however, can be challenging, as the approach sometimes conflicts with the more the methodical ways of IT. So how can people, process and technology be brought together for business success in this new paradigm?
People and Process
Digital transformation first must be defined. It’s a ubiquitous app or service that can be consumed from anywhere on any device at any time. Digital transformation aims to reduce the consumption barrier, which enables businesses to profit from every possible engagement while leveraging data-driven models.
A DevOps culture supporting such digital transformation is about agility, availability and scalability with continuous integration and continuous delivery of services. These strategies require new technology skillsets, which begs the age-old question: build or buy?
This question leads to more questions, which IT organizations struggle to answer. Should they “buy” new talent and build the capabilities with the existing team? Should they buy new talent and implode the existing team? Or, should they build up the existing team and enable them to bridge the gap from traditional ops into a DevOps culture?
Organizations often overlook one of the most important considerations. Putting top engineers with topnotch development skills together does not always produce a successful team. DevOps calls for collaboration and cooperation, so organizations must pay attention to soft skills. People collaborating across traditional boundaries and silos are at the center of digital transformation. Progress only happens when the organization is moving forward as a whole with the same end goal and execution in mind.
Digital transformation is a function of speed to market, and that’s where traditional IT and DevOps processes clash. IT follows the doctrine of trust but verify, a methodical process filled with rigor and discipline. On the other hand, DevOps’s modus operandi is continuous integration and continuous delivery while failing fast if needed, trying anything and everything to iterate and get as many data points as possible, as fast as possible.
Blending these cultures and processes is a matter of stepping back and realizing they are actually more similar than different. Striking a process balance between IT Ops and DevOps empowers organizations to deal with the volume, variety, and velocity of change involved in digital transformation.
Normalize the Right Technology
Making the right technology choices means finding the right signal in the noise of all the possibilities. From programming languages to the automation engine, infrastructure services (virtual machines, cloud instances, containers and microservices, serverless) and more, when creating a DevOps culture, people will gravitate to solutions that are biased toward what they know best.
The number of choices turns into an endless buffet with a tendency to overindulge, which creates bloat and starts accruing tech debt. Shouldering too much technology and being half-baked into “best-in-breed” solutions are not enough.
Disruptive technologies surpass the bounds of “good enough.” The wrong technology choice will put an organization behind in a world that’s disrupted constantly. The organization may get disrupted into nonexistence. Choose the technology that will best enable normal consumption of services while you scale services, become more agile, and expand availability. That’s what digital transformation done right accomplishes.
Customer-centric
A transformative DevOps culture keeps the customer as the focal point. Some organizations view this as “the customer is always right.” That’s one way to look at it, but sometimes the customer doesn’t know what “right” is. It’s up to the organizations to demonstrate good and great.
That means being less feature-centric and more solutions-centric—actually solving the customers’ pain points. Don’t get stuck in the “easy” to implement. Don’t assume your use case is the right use case without being certain of how customers would leverage and utilize the end results of the transformation. Remember, the goal is to reduce the barriers to consume for consumers.
One of those barriers is having a bad experience with an application or a service. Users want it now and are willing to pay if the service is excellent, always available as needed, and easy. We’ve seen this with rideshare applications, Netflix and Amazon.
Yet, the SolarWinds Digital Experience Monitoring Survey shows that 40 percent of respondents are suffering digital experience issues that negatively affect their businesses. They also report a lack of visibility into customer experience. Forty-four percent say end users are first to discover an issue, and 25 percent of those surveyed say they are the last to know.
Best Practices
Digital experience monitoring (DEM) tools help solve these problems. For example, when implemented as part of a rigor and discipline approach, the survey shows that 59 percent received fewer support tickets, 44 percent improved brand image, and 33 percent gained a better ability to retain customers. Among those who reported that DEM has driven increased revenue, 15 percent of revenue increase can be attributed to DEM.
Observability and Control
At the same time, a certain level of observability and control must be integrated into a successful DevOps strategy for digital transformation efforts. Observability is a measure of how well internal states of a system can be inferred from knowledge of its external outputs.
Depending on who you ask, observability can mean logging for one IT professional, metrics for another, or tracing for a DevOps engineer. Despite these shades of grey, the reality is that application stacks are evolving with so much velocity, variety and volume that such that a combination of all three (metrics, logs, and traces) are needed to build a viable observability protocol.
To achieve this level of observability—and, as a result, control—you must have a powerful integrated full-stack monitoring platform that enables analysis of key performance metrics, events and logs,and application code traces. This approach to monitoring as a discipline enables developers, DevOps engineers and IT professionals to better manage, optimize and troubleshoot the full stack at every layer—from the infrastructure services and application to the end-user experience—from a single dashboard.
With observability and controllability enabled, you can successfully and continually integrate and deliver the applications and services enabled by your digital transformation.
— Kong Yang
