By Yann Albou.
DevOps is much more than just a technological trend. It is an approach that revolutionizes the way IT teams collaborate, innovate, and respond to business needs. But is it truly the miracle solution for all businesses? In this article, we explore the fundamental principles of DevOps, its advantages, as well as its limitations and the reasons why some implementations fail.
I hope this article will help you understand the stakes of DevOps and guide you in making informed decisions tailored to your own context.
The DevOps or DevSecOps movement emerged in the late 2000s as a response to the challenges posed by collaboration between development (Dev) and operations (Ops) teams in the fields of IT and software.
There are many definitions of DevOps, but to simplify, we can say that DevOps or DevSecOps is a cultural, organizational, and technological approach aimed at improving collaboration, communication, and efficiency between development, security, and operations teams (and others). The primary goal is to reduce the “time to market” for software products, improve product quality, and respond more quickly to user needs.
It is closely tied to agile practices and is particularly a response to the issues associated with waterfall development models.
Several books have significantly influenced the history of DevOps:
Many other books could be added, especially literature on Agile and Lean practices.
DevOps practices are supported by specific tools and technologies. Here are some of the main technological approaches related to DevOps:
This list is far from exhaustive but provides a good representation of the Cloud Native landscape associated with DevOps approaches.
There are numerous tools and products related to these topics. For a detailed overview, see the CNCF landscape 😱.
As a reminder, the Cloud Native Computing Foundation (CNCF) is an organization that supports and promotes open-source technologies for cloud computing and containers. It was founded in 2015 by several leading tech companies, including Google, Red Hat, and VMware. The CNCF has become one of the most influential organizations in the field of cloud computing and containers, with over 200 open-source projects under its umbrella, including Kubernetes.
This ecosystem is incredibly rich and still not exhaustive.
It is worth noting that Sokube is a “silver” member, KCSP (Kubernetes Certified Service Provider), and KTP (Kubernetes Training Partner) of the CNCF.
These tools should be seen as building blocks that allow teams and companies to create solutions tailored to their needs and designed to align with DevOps modes of operation: automation, security, everything-as-code, and simplifying collaboration across departments.
I will not delve further into these technological aspects here.
Not at all!
Let’s start by clarifying one point: Kubernetes is not the solution to all your problems. There are many cases where Kubernetes is simply not recommended due to its complexity, cost, or its suitability primarily for medium- to large-scale environments. For certain needs, alternatives like CaaS (Container as a Service) solutions may be more appropriate.
Kubernetes is, above all, an enabler for DevOps practices. However, if you use it with poorly suited applications or replicate “legacy” approaches (in terms of delivery, security, observability, etc.), your project is likely to face challenges—or even fail.
“I use containers and Kubernetes, so I’m doing DevOps.”
I often see companies thinking they are doing DevOps simply because they use tools like containers or Kubernetes. This approach is flawed: adopting tools alone is not enough to transform your DevOps practices.
DevOps promises a real transformation: delivering faster, with better quality, while enhancing security and providing increased business value. This is often summarized by the mantra “Faster, Better, Stronger.”
However, despite all these promises of speed, quality, and security, the implementation of DevOps is not always successful. This leads us to ask a crucial question: Is DevOps a failure?
I have supported many companies in their journey to implement DevOps. However, it is not uncommon to encounter failures in establishing a truly operational DevOps model. So why do these “transformations” sometimes fail? What challenges underlie these failures?
First, let’s talk about transformation. This term, while often used, carries significant weight and can seem intimidating. It reflects decades of work methods that are no longer suited to current needs and contexts.
We are often accustomed to operating in a waterfall project management mode. This model, inherited from manufacturing industries, works in rigid, linear stages: each phase must be completed before moving on to the next. While this approach has served sectors with predictable and fixed processes, in today’s world of software and digital services, it has become problematic.
Because the digital world is fundamentally uncertain and ever-changing. User needs evolve constantly, and competition demands increased responsiveness. The waterfall model does not allow for adjustments mid-project, resulting in slow delivery that is often disconnected from actual expectations. Consequently, projects fall behind schedule, and teams end up with outdated products before they are even finalized.
True agility is not just about breaking the waterfall model into shorter sprints; it starts with focusing on creating value for the customer, gathering feedback, and solving their problems, rather than merely delivering faster.
While there are cases where the waterfall model may be appropriate (e.g., projects with very stable and fixed requirements, highly regulated environments), agility is generally preferable for continuously adapting to changes and maximizing value creation for end users.
Thus, it is essential to abandon these practices in favor of more agile and iterative methods, which promote continuous adaptation and collaboration between teams, focusing on customer needs.
The adoption of DevOps represents a promise of profound “transformation” for IT departments (DSIs). However, one of the main obstacles these departments face is organizational inertia. In such cases, lengthy and complex decision-making processes often act as a major barrier, accompanied by functional silos that hinder the collaboration and agility necessary for the effective implementation of DevOps and Agile approaches.
Modern businesses operate in constantly evolving environments, with rapidly changing needs. To remain competitive, they must transform into “antifragile” organizations—structures not only capable of withstanding constant change but also leveraging it to become stronger. Change has become the norm, and IT teams must adapt to this reality by becoming more agile, flexible, and resilient.
The concept of “Antifragile” was popularized by Nassim Nicholas Taleb in his book of the same name. It describes a system capable of adapting and strengthening in the face of change, rather than breaking or remaining static.
Additionally, everything is accelerating! Change now occurs at a much faster pace than before. The COVID-19 pandemic made us realize that we must be capable of reacting even faster.
The rise of artificial intelligence will further amplify this dynamic. Whether we like it or not, the impact of AI on our professions will be profoundly disruptive. Our IT systems must adapt to this rapid pace and also become “antifragile”!
But IT can also be a key differentiator!
“Every Company is Now a Software Company”
Satya Nadella (CEO of Microsoft)
This quote highlights an essential point: in today’s digital economy, technology and IT systems are no longer merely support functions. IT has become a key differentiator capable of delivering direct business value. Whether optimizing operations, improving the customer experience, or fostering innovation, agile and responsive IT can transform a company into a market leader.
A concrete example is Amazon, which has turned its IT infrastructure into a strategic lever. Thanks to well-established DevOps and agile practices, Amazon is capable of deploying thousands of microservices continuously, enabling rapid innovation and constant adaptation to market demands.
Smaller companies have also successfully transformed by adopting DevOps practices, including:
As we have seen, IT plays a central role in the competitiveness of businesses today, regardless of their size or sector. However, not every company is a giant like Netflix, Amazon, or Google, with massive human and financial resources. So, is it relevant for smaller organizations to adopt these approaches?
The answer is yes, but not in the same way. Even for a startup, DevOps practices bring real value. It’s not about aiming for full automation or a continuous deployment model right from the start, but rather understanding the key principles to build an IT infrastructure that is smart, scalable, and tailored to your needs. It is essential to progress step by step, depending on the maturity of your organization and your specific context.
By gradually adopting these practices, you prepare your IT to grow alongside your business. Whether driven by your ambitions, opportunities, or external constraints, DevOps allows you to support this evolution by delivering value at every stage, without rushing into overly abrupt transformations.
Starting with technology alone is not enough. The real goal is to address business needs through technology, rather than implementing technology for its own sake. Of course, technical initiatives can be launched in parallel, but it is crucial to focus on the value added to the business.
Similarly, it is essential to surround yourself with experts in agile practices and DevOps-related technologies. These specialists play a key role in spreading best practices, supporting your teams to avoid common pitfalls, accelerating skill development, and maximizing the value of your initiatives.
In a world where everything is accelerating, the gap between “low performers” and “high performers” continues to widen. To close this gap, here are some starting points:
In this journey of continuous improvement, measuring from the outset is essential. Here are some key DevOps metrics (derived from the DORA study):
These metrics alone are not sufficient; platform-level indicators are also necessary:
Finally, I would like to mention principles from The Lean Tech Manifesto by Fabrice Bernhard and Benoît Charles-Lavauzelle, which emphasize:
All these principles are not exhaustive, but they are key elements upon which you should build, embedding your organization’s unique DNA!
We are often asked: “How much does a DevOps transformation cost?” Naturally, if we’re talking about “digital transformation,” it typically involves a significant investment. This can be poorly received by the business, particularly due to the technical and organizational debt accumulated over the years.
It is therefore crucial to start in a targeted manner, focusing on specific use cases. As with software development, you can begin with an MVP (Minimum Viable Product) and iterate progressively to generate feedback on a smaller set of features.
However, it is important to maintain a “lean” platform, often referred to as a TVP (Thinnest Viable Platform), to ensure the solution remains minimalistic and easy to maintain in the long term.
That said, let’s not sugarcoat it: any transformation entails additional costs, especially since you need to manage existing systems while building expertise and absorbing new technological and organizational approaches. The key is to control these costs through successive iterations, always in service of a clear objective.
If this domain is new to you, we recommend starting with our one-day DevSecOps overview training course, which will help you become familiar with the concepts and best practices.
There is no universal solution: what works for one company cannot be identically replicated in yours, as every organization has its own specificities. That’s precisely what makes this journey so exciting! However, there are fundamental principles and concepts that must be understood and adapted to fit the structure, size, context, and goals of your company.
Artificial intelligence will profoundly disrupt the adoption of these approaches and the technologies that continue to evolve. Companies that remain anchored in a legacy mindset risk being quickly left behind.
What fascinates me about DevOps is that its principles transcend technologies: continuous improvement applies not just to projects but also to DevOps itself!