The 90/33 Gap: Why Most AI Projects Fail
Ninety percent of executives say AI is a priority. Only a third report meaningful results. The gap isn't about technology — it's about how companies approach transformation.
By John Hynds · May 12, 2026
There's a number that should make every business leader pause. According to recent industry surveys, roughly 90% of executives call AI a strategic priority. But only about a third of those organizations have seen meaningful, measurable results from their AI investments.
That's the 90/33 gap. And if you're running a mid-market business trying to figure out where AI fits, understanding why that gap exists might be the most important strategic insight you'll get this year.
The Real Reasons AI Projects Fail
After spending 40 years in enterprise technology — the last several focused specifically on AI transformation — I've seen the same patterns repeat across industries. The failures rarely come from bad technology. They come from bad approach.
1. Starting with Technology Instead of Operations
The most common mistake: buying an AI tool and then looking for a problem to solve. A manufacturer buys a chatbot platform, a field service company subscribes to an "AI-powered" scheduling tool, an insurance firm licenses a document extraction service. Each tool works in isolation but never connects to the operational workflow that actually generates revenue.
The fix is deceptively simple. Start with the operation, not the technology. Map how work actually flows through your business — from the moment a lead comes in to the moment an invoice gets paid. Identify the bottlenecks, the manual handoffs, the places where data gets re-entered or decisions get delayed. That's where AI creates real value.
2. Building Disconnected Point Solutions
The second pattern is the "tool collection" approach. A business ends up with an AI chatbot from one vendor, an automation tool from another, and some kind of analytics dashboard from a third. None of them share data. None of them understand the context of what the others are doing. The result is what I call "AI theater" — it looks modern, but it doesn't actually transform how work gets done.
Platform-based transformation solves this by building integrated systems where every component shares data, context, and intelligence. When the CRM knows what the scheduling system knows, and the estimating tool can pull from the knowledge base, you get compound returns instead of isolated improvements.
3. Underestimating the Data Foundation
AI systems are only as good as the data they're built on. And most mid-market businesses have their data scattered across spreadsheets, disconnected systems, and people's inboxes. Before you can build intelligent systems, you need connected data. That doesn't mean a massive data warehouse project — it means building a platform that centralizes the data that matters most for your operational workflows.
4. No Executive Ownership
AI projects that get delegated to IT departments without executive ownership fail at dramatically higher rates. This isn't a technology project. It's an operational transformation project that happens to use technology as its primary lever. The operations leader or the CEO needs to own it, because the decisions about what to automate, what to augment, and what to leave alone are business decisions, not technical ones.
What Successful AI Transformation Looks Like
The companies that close the 90/33 gap share a few characteristics:
- They start with a specific operational problem — not a technology demo
- They build on a platform — not a collection of disconnected tools
- They measure business outcomes — not technology metrics
- They have executive commitment — not just IT budget
- They plan for evolution — not a one-time implementation
The gap between AI ambition and AI results isn't inevitable. It's the natural consequence of approaching transformation backwards. Start with operations, build on a platform, measure what matters, and the results follow.
Where to Start
If you're wondering where your organization stands, our free AI Operations Readiness Assessment takes less than five minutes and identifies your highest-ROI automation opportunities. It won't tell you to buy anything. It will tell you where to focus.
Ready to Take the Next Step?
See where AI can make the biggest impact on your operations with our free readiness assessment.