AI Can Generate Possibilities. The Real Transformation Question Is Deciding Which Ones Matter.

Businesses are increasingly watching competitors, peers and new startups experiment with AI-enabled processes, creating pressure to determine where AI can genuinely create business value.

This growing visibility of AI is also creating pressure on organisations that continue to rely on legacy processes and systems. However, the answer to this pressure is not necessarily to introduce AI into every process. The real question is where transformation is required and whether AI is the appropriate way to achieve it.

With the rise of generative AI tools such as ChatGPT and Gemini, it has become much easier to generate a large number of possible approaches to a business problem. But the challenge remains: which possibility should be prioritised for execution?

AI can generate possibilities, but it cannot automatically validate their relevance to a particular business context. The cost of generating possibilities has gone down significantly, while the cost of implementing the wrong possibility has not.

Thus, the real transformation challenge is to understand which possibility matters, where AI can genuinely be beneficial, what should be prioritised first, and how the right priority can be turned into practical execution.

The question, therefore, is not simply how to prioritise the possible solutions generated for a business problem. It is first about understanding the context of the business problem and the outcome the organisation is trying to achieve.

Decision-makers may already consider many of these factors before making a transformation decision. However, I believe a structured evaluation should consider the following:

1. What business outcome are we trying to achieve? The starting point should be a clearly defined outcome rather than a technology or an AI use case.
2. What is the impact of solving the problem? How would solving the problem affect revenue, cost, productivity, customer experience, scalability, risk or overall business growth?
3. How frequently does the problem occur, and at what scale? A problem that occurs frequently or affects a significant part of the organisation may deserve greater attention than an isolated issue.
4. What is the actual state of the data? Is the required data structured and accessible, or is it fragmented across systems and processes? What would be required to make it usable for the proposed solution?
5. Does the solution actually require AI? The question is not, “Where can we use AI?” but rather, “What is the simplest technology approach that can produce the required business outcome?” The appropriate answer could be traditional technology, automation, AI, or a combination of approaches.
6. What is the technical feasibility and implementation complexity? A possibility may appear valuable but may require significant integration, system changes, skills, resources or time before it can deliver value.
7. What risks are involved in implementing the proposed solution? These could include operational, financial, security, compliance, data or customer-related risks, depending on the use case.
8. Is the organisation ready to adopt the required change? A technically viable solution can still fail to deliver its intended outcome if the organisation, its people and its processes are not ready to adopt the change.
9. What are the consequences of not acting? What happens if the problem remains unresolved? The cost of inaction may be reflected in lost revenue, increasing operational costs, customer dissatisfaction, scalability constraints, risk or missed opportunities.

These questions help shift the conversation from “What can we do with AI?” to “What should we do, why should we do it, and what should happen first?”

Hence, the argument is not that organisations should avoid using generative AI tools to find possible solutions to their business problems. Rather, the process should begin by determining the business context, identifying the problem we wish to solve, defining the desired outcome, using generative AI and other tools to explore possible approaches, evaluating those approaches, prioritising what matters, and then executing the right priority.

The sequence can be visualised as:

Business Context → Problem → Desired Outcome → Possible Approaches → Evaluation → Prioritisation → Execution

The distinction is important because possibility, priority and execution are three different decisions.

AI can help expand the possibilities. It does not remove the need to decide which possibilities deserve organisational attention, experimentation and investment.

AI can expand the list of things a business could do. Transformation requires deciding which of those things the business should actually do — and why, now.

Organisations therefore need a way to decide what deserves experimentation and execution based on their business context, the problem they are trying to solve, the outcome they want to achieve, and the improvements required in their operations and processes.

Not every business problem requires AI. Not every AI possibility deserves implementation.

The starting point for transformation should therefore be the business context rather than the technology.

With that thought in mind, at Novacinet we follow a four-step approach to help businesses transform:

1. Understand the problem
2. Identify what matters
3. Prioritise what should happen next
4. Turn the right priorities into practical execution

Think before you transform.