We live in an interesting time to be a CIO. The pace of technological change has become so relentless that a technology strategy decision we make today could have a significantly better, cheaper or more capable alternative within the next twelve months. Artificial intelligence, geopolitical shifts, changing business models and continuous innovation are making it increasingly difficult to predict what the technology landscape will look like even two years from now.
Yet many of us are still developing elaborate five-year ICT strategic plans, committing substantial capital to technology platforms and entering long-term vendor agreements, often on the assumption that the technologies we select today will remain competitive throughout that period.
I increasingly question whether this approach is sustainable.
Consider artificial intelligence. Imagine an organisation selecting a single AI provider as its strategic partner and signing a five-year enterprise agreement. Twelve months later, another provider releases models that are significantly more capable at a fraction of the cost, and suddenly your users are asking for access to these newer tools. The organisation now finds itself locked into a technology that may no longer offer the best value, not because the original decision was necessarily wrong, but because the market moved faster than its strategy anticipated. And here is where it gets interesting. Your users are unlikely to wait patiently for the organisation to catch up. They will simply reach for their credit cards, subscribe to the newer tools and start using them to get their work done, often outside the visibility and control of the IT function. Before you know it, shadow AI has entered the organisation, bringing with it new challenges around data protection, cybersecurity, governance and compliance. What started as a strategic technology investment has inadvertently become a business agility problem and a cybersecurity risk, all because the organisation could not adapt quickly enough.
This is precisely the kind of scenario CIOs should be preparing for.
Samsung offers an interesting example of a different approach. In June 2026, the company announced that its Device eXperience division would make ChatGPT, Google’s Gemini and Anthropic’s Claude available to employees, allowing them to use different AI tools depending on their needs rather than relying exclusively on one. While managing multiple platforms introduces its own complexities, the thinking is compelling: no single technology provider is guaranteed to remain the best at everything.
There is an important strategic lesson here. In a rapidly evolving technology market, the ability to change your technology choices may become just as valuable as the choices themselves.
Of course, not every organisation has Samsung’s resources, and this brings me to a challenge much closer to home.
Across Africa, many organisations are still operating legacy systems accumulated over decades. Some have heavily customised enterprise applications, ageing infrastructure, fragmented databases and enormous amounts of technical debt. In many cases, these systems still support critical business operations, making wholesale replacement both expensive and risky.
The temptation is often to pursue a major transformation programme that replaces everything at once. I believe this is where CIOs need to rethink their approach.
Rather than embarking on expensive, multi-year, big-bang replacements, we should increasingly focus on incremental modernisation, modular architecture and deliberately reducing technical dependencies. Where possible, decouple tightly integrated systems, introduce APIs, modernise integration layers and progressively replace components without destabilising the entire technology estate.
A legacy system does not necessarily need to be replaced in its entirety to deliver new business value. Sometimes the better investment is automating a specific business process, improving data accessibility or introducing a new capability alongside the existing platform.
This approach does more than reduce transformation risk. It preserves strategic optionality: the ability to make different technology choices in the future without having to start from scratch.
And that optionality has real financial value.
Traditionally, CIOs evaluate technology investments through total cost of ownership, return on investment and expected useful life. But perhaps we should increasingly pay equal attention to the cost of changing direction. What would it cost to exit a vendor relationship? How easily could we migrate our data? Would replacing one application require rebuilding ten others?
The cheapest technology today may not necessarily be the most economical technology over its lifetime. A slightly more expensive but interoperable platform could prove substantially more valuable if it allows the organisation to adopt better technology tomorrow.
This is why vendor portability, open standards, data ownership and sensible contractual exit provisions should increasingly become boardroom considerations, not merely technical details buried in procurement documents.
However, there is another dimension to agility that we sometimes overlook.
An organisation can invest millions in modern cloud platforms, APIs and modular architectures and still remain fundamentally inflexible because its internal processes cannot move at the same pace.
Imagine identifying a promising technology today, but requiring nine months to approve a procurement, another six months to secure a budget variation and several layers of authorisation before a small pilot can begin. By the time the organisation is ready, the market may have moved on.
You cannot build an agile technology organisation on top of rigid institutional decision-making.
This is particularly relevant across many African enterprises and public institutions, where annual budget cycles, procurement arrangements and lengthy approvals can constrain the speed of innovation. We need governance models that preserve accountability and financial discipline while creating room for controlled experimentation, smaller investments and faster decisions.
But I think there is an even bigger picture here. The purpose of building an agile technology estate is ultimately to make the business itself more adaptable. Technology agility means very little if the organisation cannot respond quickly to changes in the market, customer expectations or its operating environment.
Consider a business where fuel prices suddenly increase, pushing up distribution and operating costs. The business may need to adjust product pricing, revise delivery charges, change supplier arrangements or rethink its operating model to protect its margins. Yet in some organisations, something as seemingly straightforward as changing a pricing structure requires extensive system modifications, vendor intervention or lengthy development cycles. By the time these changes are implemented, market conditions may have shifted again.
The same applies when customer behaviour changes, a new competitor enters the market, regulations evolve or an organisation identifies a new revenue opportunity. How quickly can the business respond, and is its technology estate enabling that response or standing in the way?
This is why agility should increasingly be viewed as a business capability rather than simply an IT objective. CIOs should be building platforms that allow organisations to introduce new products, adjust pricing, automate processes, respond to regulatory changes and experiment with business models without requiring major technology projects every time something changes.
The real measure of an agile technology strategy is not how quickly IT can deploy new technology, but how quickly the business can adapt and create value when the world around it changes.
That said, I am not suggesting that everything should become short-term or that CIOs should abandon long-term planning. Critical infrastructure, cybersecurity foundations, enterprise data architecture and operational resilience require stability and sustained investment. Nor does using multiple technology vendors automatically reduce risk; it can introduce additional integration, security and governance complexity.
The real challenge is knowing what must remain stable and what must remain flexible.
Our business objectives, architectural principles and security standards can have a long-term horizon, while the technologies and delivery approaches used to achieve them should be continuously reassessed. In other words, we should be firm on the outcomes we want to achieve but increasingly flexible about the tools we use to get there.
This brings me to another capability I believe every CIO must now develop: a stronger horizon view.
The problem today is not a shortage of information about emerging technologies. We are surrounded by product announcements, predictions, AI breakthroughs and claims that the next innovation will change everything. The real leadership skill lies in distinguishing the noise from the signal.
Not every technological development deserves an investment, and not every announcement requires an immediate response. As CIOs, we must evaluate developments against business value, cost, productivity, customer experience, operational risk and long-term strategic relevance.
But when a genuine signal emerges, we must be prepared to act.
That action does not have to mean committing millions to another transformation programme. It might involve a controlled pilot, a small architectural change, experimenting with a new AI capability or modernising one component of a legacy application. The important thing is to learn quickly, measure outcomes and scale what works.
This is also why I believe the traditional five-year ICT strategy needs to evolve. We still need long-term direction, but we should resist the temptation to prescribe every significant technology choice years in advance. A strategy should clearly articulate the business capabilities we want to develop, the outcomes we expect and the investment principles that will guide us. The roadmap should remain adaptable as technology, economics and business priorities change.
Perhaps one of the most important questions a CIO should ask before approving a major technology investment is no longer simply, “Will this technology meet our requirements?”
It should also be, “How easily can we change direction if a better option becomes available tomorrow?”
And beyond technology, we should be asking an even more fundamental question: “How quickly can our organisation respond when its business environment changes?”
Ultimately, I believe the next generation of successful CIOs will not necessarily be those who make the most accurate five-year technology predictions. They will be those who build organisations capable of adapting when those predictions inevitably change.
The next competitive advantage may not belong to organisations with the most advanced technology, but to those with the greatest ability to adapt their businesses as markets, customer expectations and technologies change.
As CIOs, perhaps we should spend less time trying to predict which technologies will dominate the next five years and more time building organisations that can respond to whatever the next five years bring.
Because the ultimate purpose of technology agility is business agility. And in a world where change is inevitable, the greatest risk may not be making the wrong decision today, but losing the ability to make a different one tomorrow.
