You Don’t Have a Technology Problem. You Have a Strategy Problem.
70% of tech projects fail. The tools aren’t the issue.
Seventy percent of tech projects fail due to a lack of strategic focus rather than technical issues. Leadership often treats technology as a cost rather than a strategic lever, resulting in ineffective implementations. Successful businesses start with clear outcomes, connect investments to value, and actively manage technology adoption to drive results.
I recently came across an observation from Maurice Adam Weber that I see in boardrooms constantly and mirrors our own research what are the ingredients of a successful AI implementation project. Maurice compares tech adoption to a gym membership: we pay the fee, skip the work, and remain baffled when nothing changes.
With AI, this pattern is getting expensive.
The real failure mode
When I evaluate a business, I look past the software licenses. I look for the strategy behind them.
Most technology failures aren’t technical. They’re strategic:
A tool is purchased because competitors have it.
No one connects it to a business outcome.
Adoption stalls. The license renews. The problem persists.
This was already true for ERP and CRM platforms. With AI, the costs are climbing faster.
The mindset issue
Many leadership teams treat technology as a cost centre, not a strategic lever.
This shows up in decisions:
Technology budgets are set by IT, not by the business.
Digital transformation is delegated to a function, not owned by executives.
Success is measured by implementation milestones, not outcomes.
The result is a portfolio of tools that work technically but don’t move the P&L.
What good looks like
Businesses that extract value from technology share these traits:
They start with a business question. “How do we reduce time-to-quote by 40%” beats “Should we buy an AI tool?”
They treat adoption as a leadership problem. If the CFO doesn’t change how they work, the finance team won’t either.
They connect investment to outcomes. Every major spend has a hypothesis: what will change, by when, and how will we measure it?
They kill what doesn’t work. The best operators cut losses and reallocate.
The AI trap
AI makes this harder. The tools impress in demos. They generate excitement. And unlike traditional software, they often work in isolation but don’t create value in context.
A language model can draft a contract. Your legal costs don’t necessarily drop. A copilot can write code. Your product doesn’t ship faster. An AI agent can summarize meetings. Your decisions don’t improve.
The gap between capability and value is where AI investments die.
What to do
If you’re a company executive, stop treating IT like a utility provider. Sit down with your CTO and ask a different question.
Not: “What are we spending on AI?”
But: “Which business outcomes are we improving with AI, and how will we know it’s working?”
If the answer is vague, you’ve found your problem.