The 70% of AI Transformation Most Companies Skip

Most companies aren’t getting real value from AI.

The problem isn’t the technology. It’s how work, people, and decisions are not being redesigned around it.

Over the last year, we’ve been deep in AI transformation work: studying it, implementing it, comparing notes with peers across industries. The same pattern keeps showing up. Most organizations still treat AI like a technology upgrade, not a fundamental shift in how work gets done. They buy a platform, launch a pilot, hope the business case appears.
It usually doesn’t!

The 70% most teams skip

BCG’s latest research puts hard numbers on this: only 5% of companies are generating substantial value from AI at scale. Another 35% are making progress but admit they’re not moving fast enough. The remaining 60%? Minimal gains despite significant investment.
BCG, The Widening AI Value Gap (Sept 2025)

The gap between leaders and everyone else is widening, and it’s not because some have “better algorithms”.

The real differentiator is how work gets redesigned around AI. BCG frames it as 10-20-70: roughly 10% algorithms, 20% technology and data, 70% people, processes, and change. Most companies invert that. They overspend on tools and underinvest in operating model redesign, decision rights, and behavior change, then wonder why results stay incremental.

Deloitte’s 2026 State of AI report shows the same disconnect: 74% of organizations want AI to grow revenue, but only 20% are seeing it happen. And while 74% plan to deploy AI agents within two years, only 21% have mature governance to manage them.

Deloitte, State of AI in the Enterprise (Jan 2026)

Companies are deploying faster than they’re redesigning. That’s the gap.

McKinsey reinforces the point from their work on the agentic organization: as humans and AI systems start working together, the operating model becomes the strategy. AI changes how work is orchestrated, not just how fast tasks get done. Their research estimates 75% of current jobs will require redesign, upskilling, or redeployment by 2030, and that without deliberate shifts in workflows, governance, and culture, even the most advanced AI systems will stall.

McKinsey, The Agentic Organization (Sept 2025)

The hidden variable nobody wants to name, identity

There’s also a human dimension that gets underestimated, professional identity.
When AI can draft, analyze, summarize, and recommend at machine speed, people quietly start asking real questions. What’s my contribution now? What decisions am I still accountable for? What makes me valuable?

When leadership doesn’t address those questions directly, adoption stays superficial, shadow AI grows in the background, and trust fades away. This isn’t a “change management” checkbox. It’s the difference between tools people tolerate and capability people own.

There are also norms that need to be deliberately embedded if any of this is going to work in the long run. Culture has to evolve so people see AI as a partner in how they create value, not as a threat; security awareness has to keep pace with new attack surfaces and risks created by AI; and ethical use can’t be a slide in a training deck, it needs to be part of how products are designed, how data is handled, and how performance is measured. If we ignore these fundamentals, we might see some short-term wins, but we won’t build sustainable value at scale.

Why leaders hesitate, and what they actually need

Many executives hesitate to invest in AI because it looks expensive and risky in the short term, and the upside is often framed in vague, long-term language. In conversations with peers and people I’ve met through my courses, a common theme is that leaders rarely get a proper, structured assessment of where AI can genuinely create ROI; they’re not offered an effective workshop that tackles the tough questions across strategy, operating model, people, culture, security, and ethics, and quantifies whether this is a good investment or not. Without that rigor, AI feels like a cost, not a strategic lever.

This is why I believe AI transformation must be treated as an operating model problem, not a tool rollout. You start with a factual baseline: where are we on data, technology, governance, workforce capability and risk posture. You tie AI directly to business outcomes and specific workflows, not just high-level “use cases”. You design governance that is practical and embedded in day-to-day decisions, so executives can sponsor with confidence and teams know what responsible use actually looks like.

There’s also a practical starting point that deserves more attention: Business Readiness. Before you can redesign workflows around AI, you need to know your workflows. Organizations that have clear ownership of their processes, SOPs, and decision points move fastest because AI has something concrete to attach to. But knowing your workflows isn’t enough on its own. The trap is chasing quick wins that look impressive in a demo but don’t solve problems people feel. I see this across industries: dozens of AI features get built, but the only ones customers pay for are that eliminate real pain in a specific workflow. Start where the friction is clearest, prove value there, and use that momentum to fund the bigger transformation.

Quick wins aren’t just nice to have; They’re how you earn the right to change at scale.

Three questions that cut through the noise

If your organization is running pilots but struggling to move beyond experimentation, start here:

1. What outcome are we solving for, and which workflow needs full redesign to deliver it?
2. Who makes key decisions, and how do we measure what’s working?
3. What skills and norms need to change for adoption to stick at scale?

How we approach this at 7Sparx

This is exactly the work we do.
At 7Sparx, we help organizations move from AI experimentation to execution through a structured path.

Enterprise AI | AI Training & Enablement | 7Sparx

In Conclusion

Many leadership teams right now are stuck between experimentation and execution, and that’s normal. It’s where the real learning happens.

The organizations that break through aren’t the ones with the best algorithms. They’re the ones willing to redesign how work gets done, how decisions are made, and how people create value alongside AI. That’s the 70% most companies skip, and it’s where sustainable impact actually lives.

If you’re ready to embark on your AI journey, we encourage you to explore our specialized AI courses or book an AI consultation with our experts to assess where you are at today.