Most AI projects fail. But the best ones follow these 3 rules.
Daniel Jacobs
IT Strategy for High-Stakes M&A | Faster M&A, Smoother Exits | De-risking IT & unlocking value in M&A deals
Last week, at a leadership roundtable in London, a CEO broke through the noise with a single, incisive question: “Show me where AI is truly working.”
It posed a reasonable challenge. AI investment is at an all-time high—projected to exceed?￡500 billion by 2025 (IDC, 2024)—yet the outcomes remain frustratingly elusive for numerous companies.?Where are the genuine gains? Who is successfully implementing AI? And why are some companies experiencing 2.5 times revenue growth while others struggle to break even?
After analysing dozens of successful AI implementations across various industries, I’ve uncovered a distinct pattern—one that distinguishes AI leaders from laggards.?The winners are not those chasing the latest models; instead, they see AI not merely as a technology project but as a behavioural revolution.
The Three Patterns of AI Success
Looking beyond the glossy AI case studies, you will discover a surprisingly consistent formula for success. AI-driven organisations concentrate on three principles:
1. Start Small, Think Big
Many companies adopt a “boil the ocean” approach—embarking on ambitious AI initiatives without clear, measurable outcomes. However, the most effective AI transformations commence with?targeted, high-impact problems.
Consider?UPS’s ORION system. On the surface, it’s a route optimisation algorithm. However, beneath the hood, it’s an AI-driven revolution in logistics. ORION?saves UPS 10 million gallons of fuel annually, reducing costs while enhancing sustainability.
Could you tell me why it succeeded?
? It solved a specific problem (route inefficiencies).
? It leveraged existing data (GPS, driver logs).
? It had a clear success metric (fuel savings).
Contrast this with companies that implement AI without a clear purpose, only to find themselves overwhelmed by complexity and unproven ROI.
Takeaway:?Identify?one high-value process?to optimise with AI—one for which data is already available, the impact is measurable, and teams are prepared to embrace change.
2. Build Trust Before Scaling
AI does not fail due to imperfect algorithms; it fails because of?human resistance.
68% of consumers are concerned about AI privacy and data usage (KPMG, 2024).?Employees are apprehensive about automation replacing their jobs. Leaders face challenges in effectively communicating AI’s value to encourage adoption.
Winning organisations take a different approach. They embed trust into AI from day one:
?? Privacy is a design principle, not an afterthought.
?? Transparency about AI-driven decisions.
?? Empowering employees with AI rather than replacing them.
Consider?NatWest’s AI-driven fraud detection system. Instead of merely implementing an AI model, NatWest involved customers from the outset—clarifying how AI would enhance security without jeopardising privacy. The outcome??Increased customer trust, better fraud detection accuracy, and fewer false positives.
Takeaway:?Adopting AI is not merely a technical challenge—it’s a psychological one.?Articulate the ‘WHY’ before presenting the ‘HOW’.
3. AI Success is a People Strategy, Not a Tech Strategy
The most successful AI-driven companies are not those with the best algorithms; instead, they are the ones that invest the most in their?people.
According to?Accenture (2024), organisations that invest in AI upskilling experience 2.5 times the revenue growth compared to those that do not.
Why? Because?AI provides value only when humans understand how to use it.
Consider?Coca-Cola’s AI-powered marketing transformation. The company didn’t merely implement AI for digital ad optimisation; instead, it cultivated internal AI fluency among marketing teams, ensuring that employees could interpret AI-driven insights and adjust campaigns in real time. The outcome??Increased ROI, enhanced customer engagement, and a lasting competitive advantage.
Takeaway:?AI isn’t a magic bullet.?Invest in training, upskilling, and cultural adoption—because even the most powerful AI is ineffective if your people don’t trust or comprehend it.
Your 30-Day AI Action Plan
For IT leaders seeking to navigate the AI hype and achieve tangible results, here is a practical roadmap:
Week 1: Assessment
? Map your organisation’s data assets.
领英推荐
? Identify high-friction processes that AI could improve.
? Survey teams to assess AI readiness and capability gaps.
Week 2: Quick Wins
? Choose one process to automate or enhance with AI.
? Identify a small, motivated team to pilot the initiative.
? Define clear success metrics (cost savings, efficiency gains).
Weeks 3-4: Implementation & Iteration
? Launch your AI pilot—but keep it focused.
? Gather daily feedback from users.
? Adjust based on real-world learnings before scaling.
Interactive: Rate Your AI Readiness
Is your organisation prepared to move beyond the hype of AI and achieve tangible results? Rate yourself on the following:
Data Quality
? Do you have structured, accessible data?
? Is data governance in place?
Team Capability
? Do you have internal AI champions?
? Is leadership aligned with AI priorities?
Business Case
? Have you identified clear AI use cases?
? Can you measure potential impact?
?? 0-3 points: Focus on AI foundations.
?? 4-6 points: Ready for pilots.
?? 7-9 points: Scale existing AI initiatives.
The AI Gap is Widening—Where Will You Be?
By 2025, the divide in AI will shift from a distinction between adopters and non-adopters to one between those who utilise AI effectively and those who do not.
The winners will be organisations that?view AI as a behavioural challenge—not merely a technological one.
So, here’s my question for you:?What is the one process in your organisation that could be transformed by AI within the next 90 days?
Drop a comment below—I’d love to hear your thoughts. ??
?? Lisa Rabone - Sustainability in Data (SiD) AI is definitely more than just tech. I don’t think there is a precedent to AI
I think companies who ask that question do not understand AI deployment correctly. They think it’s a silver bullet that provides results immediately, whereas in fact, integrating AI solutions into your business is just about working with the AI from a determined baseline and scaling up towards the results you want to achieve. AI is like a member of staff, it gets things wrong and can take time to get right. It needs attention to bed in and be integrated— to understand your business logic and policies, just like any new member of staff would do. With AI, you get what you put in, but it’s almost RARELY an immediate silver bullet. However, if you have the foresight to stick with and understand the AI, what it does, how it works and what its potential can do, then you will only go upwards with AI, and exponentially with that.
Automation Advocate |No code Automation Solutions| Driving Business Efficiency with Affordable Automation at Msquare Automation Solutions | Digital Transformation | Go to Market Strategies | Marketing |AI integration
2 周Daniel Jacobs A first hand experience for me. Started without identifying the specificity of the problem.
? Prompting AI, Leaders & Industries to Think Differently | Keynote Speaker? Director of Data & ESG Strategy | Strategic Advisor ? Collaborative Innovator ? Dyslexic Creative Leader
2 周Haven’t we been here before? The AI gold rush feels eerily familiar—another round of chasing the next shiny, all-solving technology without due diligence. The key patterns outlined here are solid, but history shows us that tech alone doesn’t guarantee transformation. We’ve seen this cycle with big data, blockchain, and countless other ‘revolutions.’ The real question is: Are we learning from past missteps? AI success isn’t about tools; it’s about mindset, culture, and strategic alignment. Until organisations embed AI as a long-term behavioural shift—rather than another box-ticking exercise—the gap between hype and impact will persist.
I help Executives, HR and Leaders build High Performing Teams & Organisations by enhancing Engagement, Effectiveness and Empathy | 3x Author | High Performance Expert??
2 周Daniel Jacobs This is spot on! AI success is definitely more about people than just technology.