Navigating the cloud maze: A 5-phase approach to optimizing cloud strategies

In today’s fast-paced digital landscape, the cloud has emerged as a cornerstone of modern business infrastructure, offering unparalleled scalability, agility, and cost-efficiency. As organizations increasingly migrate to the cloud, however, CIOs face the daunting challenge of navigating a complex and rapidly evolving cloud ecosystem. With the AI revolution underway — which has kicked the wave of digital transformation into high gear — it is imperative for enterprises to have their cloud infrastructure built on firm foundations that can enable them to scale AI/ML solutions effectively and efficiently. Otherwise, companies will struggle to realize business value with AI/ML capabilities left to endure high cloud cost expenses, as it has been for many companies in 2024 for AI solutions.  

The five-step process explained below will help enterprises and CIOs to embark on the cloud journey, which is strategic with the right provider that fits their needs, build it right the first time not go back and fix it later, inculcate the mindset of how to make money with cloud, drive awareness and train business and IT on cloud financials that should overall drive a culture of architecting cloud cost-effective and scalable solution designs that becomes a organizations’ DNA.  

The journey I propose is in five phases through the lens of a crawl-walk-run framework.  

1. CRAWL: Design a robust cloud strategy and approach modernization with the right mindset  

Modern businesses must be extremely agile in their ability to respond quickly to rapidly changing markets, events, subscriptions-based economy and ‘excellent experience’ demanding customers to grow and sustain in the ever-ruthless competitive world of consumerism. The ability of cloud infrastructure to adapt, scale and be malleable to meet business needs right where they are and then enable them to expand makes it the default choice. With the advent of growing AI adoption, a strong cloud foundation pillar is a prerequisite.  

Key considerations for cloud strategy and modernization  

• The ‘what’: The executive leadership team of business and IT together need to evaluate business needs and their current business challenges, global footprint and current technology landscape and define the company’s “Northstar,” (aka, the “what,” the vision). 
• The ‘why’: Establish clear business principles to address why you’re doing this. For instance, to deliver faster and improved quality of service to customers, faster deployment of new products and services, to increase revenue and profit margins while controlling costs, sustainability or market expansion, to name a few. 
• An enterprise with a strong global footprint is better off pursuing a multi-cloud strategy. It prevents vendor lock-in, gives a lever for strong negotiation, enables business flexibility in strategy execution owing to complicated architecture or regional limitations in terms of security and legal compliance if and when they rise and promotes portability from an application architecture perspective. In some cases, the business domain in which the organization operates (ie, healthcare, finance, insurance) understandably steers the decision toward a single cloud provider to simplify the logistics, data privacy, compliance and operations. 
• Technology modernization strategy: Evaluate the overall IT landscape through the lens of enterprise architecture and assess IT applications through a 7R framework. (rehost/lift & shift, replatform, replace, refactor, rearchitect, retire, retain). The assessment provides insights into the current state of architecture and workloads and maps technology needs to the business objectives.  

The first three considerations are driven by business, and the last one by IT. Enterprise architecture is the bridge capable of tying the two loose ends, breaking down silos and driving cloud transformation and execution strategy effectively. This new paradigm of the operating model is the hallmark of successful organizational transformation. All the major cloud providers from North America — AWS, Google, Microsoft Azure, Oracle Cloud — are on par with each other, with most of their services and capabilities are primed to address the needs of any enterprise. 

Having said that, there are a couple of standouts I would like to point out. First, cloud provisioning through automation is better in AWS — CloudFormation and Azure — Azure Resource Manager compared to the other cloud providers. For larger enterprises with a multi-cloud strategy, Terraform is the best way to go to close that shortcoming. 

Second, from a cloud security perspective, Azure has a comprehensive cloud security tool that comes as a package and can seamlessly be integrated with its services, thus enhancing the cloud security posture without the need to buy an external tool. 

2. WALK: Establish a strong cloud technical framework and governance model 

After finalizing the cloud provider, how does a business start in the cloud? You would be surprised, but a lot of companies still just start without having a plan. How difficult can it be, after all? You get a subscription and begin deploying resources. That probably works fine with really small environments, but you will soon discover that it grows over your head. You are heading for a cloudshock. No joke. 

Well-architected frameworks were henceforth invented to help customers build environments in public clouds by providing them with best practices. The good news is all major cloud providers frameworks do the same thing: 

• Operational excellence 
• Security 
• Cost optimization 
• Reliability 
• Performance efficiency 
• Sustainability

The framework helps in implementing the financial controls (FinOps) that we will discuss separately, management of workloads (BaseOps) and security controls (SecOps). At this point in the journey, the timing is right to set up a cloud Center of Excellence (CoE) practice, join hands with cloud engineers, architects and the DevSecOps community and publish the customized version of a well-architected framework that fits the individual company needs and establishes a standard artifact. This serves as a baseline to drive cloud adoption and getting started with a shifting mindset on how to ‘make money with cloud’ and not just cost savings alone. 

It’s a good idea to establish a governance policy supporting the framework. This includes the creation of landing zones, defining the VPN, gateway connections, network policies, storage policies, hosting key services within a private subnet and setting up the right IAM policies (resource policies, setting up the organization, deletion policies). Creating awareness of the policy of least privilege and addressing frustrations when cloud users ask for more to play with, and as a cloud CoE team, you are rightfully holding your ground that comes with it. The cloud CoE team should collect feedback from the users and tweak policies along the way as they deem fit. As the enterprise user community matures through this learning curve, the CoE team has a pivotal role to play in engaging them proactively, supporting them, meeting their needs and helping them address their pain points to lay a strong foundation for building a robust cloud infrastructure that is scalable to deliver business value. 

Partnering with the enterprise architecture team in this stage of the cloud journey can speed up cloud maturity and buttress the foundation further on solid bearings. The cloud CoE team of architects should work with the EA to align with the reference architecture patterns that the CoE team would like the application teams/product teams to follow in their solution design. This strategic and collaborative work serves as a blueprint for the architects to ‘show and tell’ the concept of designing cloud-effective solutions and shape mindsets towards the ‘making money with the cloud’ vision. Every company that wants to succeed in scaling AI solutions and capabilities today needs to get this design thinking working for them. Otherwise, it’s like the story of ‘sour grapes’ with high cloud cost bills coming their way with little value added in the AI domain as it has been for a lot of companies in 2024. 

3. WALK FASTER: Develop cloud adoption planning and migration roadmap  

With the CoE and EA teams working in tandem, it’s time to engage the business stakeholders/product owners and develop the plan and roadmap. Understanding business constraints and priorities, and evaluating domains that have opportunities to expand their revenue and grow exponentially upon moving to the cloud are the key parameters to keep in mind. I recommend a structured approach for this phase of the journey. 

• Assessing application architecture: Assessing application architecture is the first step in migrating to the cloud. This involves identifying which components can be lifted and shifted directly to the cloud and which might require re-architecture for cloud optimization. The AWS Cloud Adoption Framework (CAF) is an effective tool that helps to evaluate cloud readiness. 
• Identifying migration strategies and tools: Identifying the right migration strategy and tools is pivotal for transitioning legacy applications into the cloud. The 7Rs of cloud migration, which stand for relocate, re-host, re-platform, refactor, repurchase, and retire, include the standard patterns for cloud migration. The choice of strategy depends on the state of the workload. 
• Planning and execution: Planning meticulously and executing the migration process with precision is crucial for a successful migration. This involves selecting the most suitable migration strategies and tools, planning to ensure minimal disruption, and executing the migration process with precision. 
• Modernization: Modernization modifies the existing application to make it work better in the cloud. This can involve refactoring or re-architecting the application to take advantage of cloud-native services. 
• Best practices: Leveraging the best practices for smooth transition, compliance and tools of the respective cloud provider adopting a phased approach with the support of cloud partners is recommended. Investment in training and change management is critical to the success.  

4. RUN: Configure and establish a strong cloud financial model  

The centerpiece for setting up the cloud the right way for the first time and realizing the true meaning of ‘making money with the cloud’ hinges on developing a strong cloud financial model. The effectiveness of the cloud CoE comes into play in this phase. The cloud CoE team must partner with the finance team and design an overarching organizational structure within the cloud that aligns with the company’s financial reporting setup. They must then explain the ‘Pay as you go’ model clearly and the blind spots that come along with it, as the infrastructure cost now shifts from capex to opex for the most part from a portfolio management perspective. 

In my personal opinion, the “Pay as you go” model is a double-edged sword that can lead to tardiness around cloud financials, or if done right, lead to ‘making money with the cloud’. The CoE in this phase has to proactively engage ALL stakeholders, making them aware of the change in the mode of operation and guiding the team through training, coffee chats, and CoE sessions during this period of transformation. 

Cloud cost visibility, cost insights, cost governance, defining a cloud baseline infrastructure and vendor management are essential components of a comprehensive cloud financial management strategy. Cloud cost visibility involves establishing robust processes and tools to gain comprehensive visibility into cloud costs, while cost insights involve extracting meaningful insights from cloud cost data to identify trends and patterns, setting up a baseline infrastructure that can be rinsed and repeated. The baseline infrastructure serves as a ballpark estimate of the financial cost that can be modeled for the rest of the usage. 

A critical process that is a must for companies to get started on the right footing is having the cloud CoE partner with cloud architects to develop a baseline infrastructure for compute, storage, database, data transfer, application monitoring and logging services that shall be provisioned for all requests. The message has to be clear enough to the stakeholders that cloud provisioning is not a blank check. This process of governance empowers the CoE team to engage those particular stakeholders who need more than the baseline infrastructure provisioned, have them justify the need for more, and then approve changes to their request upon evaluation. 

This rule of engagement lays the foundation for building a strong inherent FinOps culture within the IT and business teams preemptively, rather than being exposed post-go-live to an expensive cloud bill. It has been identified that compute, storage, database, application monitoring and logging services drive up 75% of the cloud bill and are major outliers of cloud cost spiraling out of control for organizations. 

5. RUN FASTER: Mature the enterprise cloud usage and optimization through a strong FinOps culture  

The beauty of cloud usage is that there are always opportunities to leverage cloud services lean and mean. First, the ‘mean’ part. Due to the inherent competitiveness of various cloud providers to expand ruthlessly, they are constantly working on cutting-edge improvements and optimization features for their services that will be made available to their customers. It is indeed an overwhelming affair to keep track of new embellishments coming your way. However, the enterprises that are perfectly attuned to a strong FinOps culture are poised to take advantage of which of those ‘freebies’ thrown at them serve their needs and quickly realize their benefits. 

Because of this aspect of taking advantage of the right cloud optimization services or ‘perks’ thrown at customers, the scalability of solutions is very much possible cost-effectively, and that covers the ‘lean’ part. Cracking this code or aspect of cloud optimization is the most critical piece for enterprises to strike gold with the scalability of AI solutions. Companies must get this puzzle solved right to avoid the disappointing ROI that many have experienced in 2024 related to their AI capabilities.  

Cloud cost optimization involves identifying areas of overspending, rightsizing resources, understanding how to effectively use prompt engineering techniques and the right LLM models within AI, and leveraging pricing models and discounts that cloud service providers (CSPs) offer. Strategic metrics and criteria should be established to incorporate sustainability goals into various FinOps capabilities, and engineering and product teams should take responsibility for cloud usage, making appropriate choices in architecture, system design, license use and operational features. 

The fundamental shift in the mindset is that ‘cloud cost management’ is a shared responsibility of cloud architects, product owners, business partners, application developers leveraging DevSecOps and automation pipelines, and data teams all involved with cloud usage for a particular application. 

This point in the journey is where enterprises have achieved what I feel is the true essence of ‘making money with the cloud’. Enterprises are empowered to run faster and go big with their innovations, and not be alarmed to receive cloud shocks at the month-end with outrageous cloud bills that dissuade CIOs from moving forward because they are firmly footed with the right tools and practices that enable them to catch anomalies proactively, alter configurations, and course-correct swiftly in their cloud infrastructure or make a decision if the business use case is worth pursuing. 

Santhosh Gottigere is an IT Technology leader, who has led major business and IT transformations in enterprise architecture, service operations, human resources, coud migrations, and SAP ERP spanning across multiple vertical industries such as healthcare, pharmaceuticals, energy and gas, retail and manufacturing, consumer goods, and transportation and freight management across the US, Europe and Canada. He is a global information technology leader with broad experience from multinational IT consulting organizations and leading independent software vendors. 

This article was made possible by our partnership with the IASA Chief Architect Forum. The CAF’s purpose is to test, challenge and support the art and science of Business Technology Architecture and its evolution over time as well as grow the influence and leadership of chief architects both inside and outside the profession. The CAF is a leadership community of the IASA, the leading non-profit professional association for business technology architects.