I’ve already dedicated a lot of words to “you don’t need an AI Strategy you need a business strategy” and after a recent networking event where I feel like I said that in almost every one of the connects, I thought I’d dedicate some more time to the topic, and create something actionable.
In recent years a narrative seems to have taken over, the idea that simply adding Artificial Intelligence to a business will magically solve problems and guarantee success. This view is fundamentally flawed. It’s easy to understand why people fall for the hype. We all crave quick fixes, and AI promises to lighten our load. However, experience (and data) shows that relying on AI alone is a costly mistake. AI is a powerful tool, but it is not a business strategy in itself.
One major misconception is viewing AI as a one-size-fits-all miracle solution. In reality deploying AI without clear purpose or planning often leads to disappointment. It’s crucial to temper expectations, AI requires careful integration into your business processes to deliver true value
When companies rush in, they soon learn that AI isn’t a magic wand. For example, Gartner research found that by the end of 2025, 30% of generative AI projects will be abandoned at the proof of concept stage due to issues like poor data quality, inadequate risk control, or lack of clear business benefit.
Pouring resources into AI technology without fixing underlying strategy or process issues doesn’t solve anything, My employer has spent the best part of 5 years working on data modernisation, what we now call Data Products without this work we would be nowhere when it comes to AI. Today we’re able to move at speed.
Simply throwing money and resource at AI is like buying a booking a Waymo car, but with no set destination in mind. When organisations focus exclusively on AI technology while ignoring fundamental business strategy and processes, they set themselves up for failure.
If your processes and goals aren’t clear or are outdated, an AI system will struggle to produce useful results. I’ve seen cases where firms invested in fancy algorithms only to discover their data was scattered and their objectives undefined, the result? The AI simply amplified the confusion.
Another fallacy is the belief that “more data” automatically means better AI. Firms sometimes chase massive data collection in hopes of striking gold, yet neglect the quality and relevance of that data. This data overkill trap has been noted as a common pitfall, pouring in heaps of data without proper analysis or understanding can lead to inaccurate conclusions and wasted effort.
In my experience, a smaller set of well-curated data aligned to clear questions beats a data swamp with no strategy. AI doesn’t inherently know which data matters, that guidance must come from a sound data strategy (which we’ll discuss in Part 2). For example if you want to deploy a Contact Centre agent, don’t show it examples of bad, only show it good and it will only know to do good. Introducing bad calls into its fine tuning will create unexpected behaviour.
The core problem with the “AI will solve everything” mindset is that it ignores the essential foundations of success, a guiding business vision and a strong data backbone. Without those, AI is adrift. One Forbes Technology Council CIO put it succinctly:
AI is a powerful tool, not a silver bullet, and it must be guided by thoughtful planning to realize its value
Treating AI as a quick fix leads to what I call “random acts of AI” pilots and systems that don’t ladder up to any larger goal. I recall a conversation with a fellow product owner who was frustrated that their expensive AI customer chatbot wasn’t delivering “flawless” service out of the box. We discussed how success with AI is a journey of continuous improvement, not a one-time install. The project was restarted with improved data - as I mentioned above - and then a phased rollout, training the model gradually and refining it with feedback, proved far more effective than expecting immediate perfection. This example underscores a broader lesson put simply; implementing AI requires iteration, stakeholder education, and realistic goals, all of which tie back to strategic planning.
To sum up Part 1: The view that AI alone can drive business success is misguided. AI must be treated as one tool in the toolbox, in service of a clear strategy, It is not and can never be an all-encompassing strategy itself. Companies that over-emphasise AI without laying proper groundwork often face costly setbacks, from unmet ROI to systems that no one uses.
In the next section, I’ll argue that the antidote to this problem is going back to basics, establishing a robust business strategy and a thoughtful data strategy before doubling down on AI. Far from diminishing AI’s importance, this approach ensures that when we do apply AI, it truly supports and amplifies our goals (rather than aimlessly spinning its wheels).
Note Parts 2 and 3 are partially paywalled, if you get value from these articles please consider becoming a paid subscriber to help support my writing.