Your board approved the AI budget. Your strategy deck got standing ovations. Your teams are running pilots across six departments.
And you're still waiting for the ROI to show up.
You're not alone. KPMG just surveyed 2,110 C-suite leaders across 20 countries and 8 sectors, and the gap between AI ambition and AI results is staggering: 95% of companies have an AI strategy and 39% are scaling adoption across the enterprise, but only 8% report tangible return on investment.
That's not a rounding error. That's a $186 million problem — the average amount companies expect to spend on AI in the next 12 months.
The Numbers That Should Worry Every CFO
KPMG's Global AI Pulse Q1 2026 asked the hard questions about AI maturity, and the answers reveal a massive execution gap across enterprise AI.
Here's what's happening:
- 95% of companies have an AI strategy (up from ~70% in 2024)
- 39% say they're scaling AI or driving adoption enterprise-wide
- 52% are using AI to automate workflows across functions
- Only 8% have seen tangible ROI from those investments
The investment side looks equally aggressive:
- Average AI spend expectation: $186 million over the next 12 months
- 58% prioritize IT infrastructure upgrades
- 50% are boosting cybersecurity and data protection for AI workloads
So money isn't the problem. Access to technology isn't the problem. Willingness to experiment definitely isn't the problem.
The problem is execution at scale — and specifically, how enterprises are (or aren't) structured to support AI across the entire organization.
What Separates the 11% AI Leaders From Everyone Else
KPMG identified 11% of organizations as "AI leaders" — companies that demonstrate the ability to translate AI investments into measurable business outcomes at scale.
Here's what they're doing differently.
They create agent ecosystems in an orchestrated way. Instead of getting stuck in pilot purgatory, AI leaders deploy multi-agent systems that genuinely transform business outcomes across workflows. The difference shows in the data: only 9% of all organizations have orchestrated multiple agents across workflows, but leaders are doing this at significantly higher rates.
In practice, this means a Fortune 500 manufacturer doesn't just pilot an AI agent for demand forecasting in one region. They deploy coordinated agents across procurement, inventory optimization, logistics routing, and supplier communication — and those agents share context, escalate decisions appropriately, and operate within defined governance boundaries.
They upgrade governance to manage risk and preserve trust. This is the part most organizations get backward: they treat governance as an afterthought or a compliance checkbox. AI leaders build governance directly into how agents are designed and deployed from the beginning.
The data backs this up: 81% of AI leaders say they have the capabilities and governance to manage AI risk at scale, compared to only 63% of non-leading organizations. Leaders also report higher board-level AI expertise and stronger integration between compliance, cybersecurity, and AI deployment workflows.
For CIOs, this means clear ownership of AI-driven decisions, real-time monitoring and observability built into system architecture, and adaptive controls that keep pace as agents scale across teams. As Samantha Gloede, KPMG's global head of risk services and global trusted AI leader, puts it: "When you start coordinating multiple AI agents across business functions, getting the governance right is both difficult and vital. CIOs need to be clear about who owns decisions made by agents because once agents operate across teams, decisions don't sit in one place anymore."
They bring their people with them. AI leaders don't just deploy technology — they invest in workforce readiness, support teams through organizational change, and build the skills needed as AI becomes part of everyday work.
This is where most enterprises are falling short. Only 22% say they're "very confident" their talent pipeline can meet the needs of an AI-enabled workforce, and 25% identify workforce readiness as a top challenge. Technology isn't the bottleneck. People are.
The ROI Gap: Leaders vs. Laggards
The business value difference is stark.
82% of AI leaders say they've seen meaningful business value from AI investments, compared to 62% of organizations still piloting. Leaders are also 2.5 times more confident in their ability to manage AI risk than non-leaders.
What does "meaningful business value" actually look like? In conversations with enterprise leaders implementing agentic AI at scale, here's what the ROI breaks down to:
Cycle time reduction: A global bank reduced loan underwriting from 4-6 days to 8-12 hours by orchestrating AI agents across credit scoring, document verification, fraud detection, and compliance checks. That's not just faster — it's a competitive advantage when a commercial client can get financing approved same-day instead of waiting a week.
Cost per transaction optimization: A retailer cut customer service costs by 35-40% through AI agent triage and resolution, handling 70% of tier-1 inquiries end-to-end without human escalation. On a $50M annual support budget, that's $17-20M in savings while maintaining or improving customer satisfaction scores.
Revenue acceleration: A SaaS company deployed AI SDR agents that identified qualified leads, personalized outreach, and scheduled discovery calls with 40% higher conversion rates than manual SDR workflows. For a sales team targeting $100M ARR, that incremental 40% conversion improvement translates to millions in additional bookings.
But here's the critical distinction: leaders didn't achieve these outcomes by running isolated pilots. They achieved them by redesigning workflows, governance structures, and team responsibilities to support AI operating at scale across the enterprise.
The Governance Challenge No One's Talking About
Most organizations say they're hampered by governance complexity, and the data shows why.
While 52% claim they're using AI to automate workflows across functions, only 9% have successfully orchestrated multiple agents across those workflows. The gap is governance: fragmented systems, unclear ownership, and lack of real-time observability as agents make decisions across team boundaries.
Here's the operational problem: when a single AI agent operates within one team (say, demand forecasting in supply chain), governance is straightforward. You know who owns the model, who validates outputs, and who intervenes when predictions look wrong.
But when you coordinate agents across procurement, inventory, logistics, and supplier communication, governance becomes exponentially more complex. Who owns the decision when the procurement agent wants to order 10,000 units based on demand forecasting, but the inventory agent flags warehouse capacity constraints, and the logistics agent predicts a 3-week shipping delay?
AI leaders solve this with clear ownership hierarchies, real-time monitoring dashboards, and adaptive controls that escalate decisions appropriately. Non-leaders either avoid multi-agent orchestration entirely (staying stuck in pilots) or deploy agents without sufficient governance and deal with the operational chaos that follows.
For CTOs and CIOs, Gloede's advice is direct: "Don't treat governance as an afterthought. Build it into how agents are designed and run from the beginning. That means ownership, accountability, and controls are clear as agents move across teams and functions."
The Workforce Readiness Problem
Even when companies nail the technology and governance, they hit a third roadblock: their teams aren't ready.
Only 22% of organizations are "very confident" their talent pipeline can meet AI workforce needs, and 25% identify workforce readiness as a major challenge. This isn't about hiring more AI engineers or data scientists (though that helps). It's about upskilling existing teams so they can work effectively alongside AI agents.
What does workforce readiness actually mean in practice? It's the difference between a sales team that views AI SDR agents as a threat to their jobs versus a sales team that uses AI agents to handle prospecting and qualification, freeing them to focus on high-value relationship building and deal negotiation.
It's the difference between a finance team that manually reviews every AI-generated forecast versus a finance team that understands model assumptions, knows when to trust predictions, and escalates edge cases appropriately.
AI leaders are investing in hands-on learning embedded into real workflows rather than abstract training sessions. They're launching sandbox environments where teams can experiment with AI tools in real-world scenarios, running internal competitions that reward employees who build AI solutions with measurable business impact, and creating cash prize incentives for solutions that deliver client value or operational improvements.
KPMG uses this approach internally, and Gloede reports it's effective: "Approaches like these help us build a workforce that can adapt and thrive as our profession evolves."
For enterprise leaders planning AI deployments, workforce readiness should be a parallel workstream from day one — not something you address after the technology is deployed.
What This Means for Enterprise Buyers
If you're evaluating AI vendors, planning deployments, or justifying next year's AI budget, here's the strategic read.
Most enterprises are spending aggressively but not seeing returns. The average company expects to invest $186 million on AI over the next 12 months, but only 8% report tangible ROI. That's a massive capital allocation risk if your organization falls into the 92% that's spending without measurable outcomes.
The winners aren't just piloting — they're orchestrating. AI leaders are deploying multi-agent systems across workflows with clear governance, real-time observability, and defined escalation paths. If your AI strategy is still focused on isolated pilots, you're already behind.
Governance is the unlock, not the bottleneck. Leaders build governance into system architecture from the beginning — ownership, accountability, and controls are clear before agents deploy, not retrofitted after chaos emerges.
Workforce readiness is as critical as technology. You can't buy your way out of the skills gap. Leaders invest in hands-on learning, sandbox environments, and incentive structures that reward teams for building AI solutions with real business impact.
The execution gap is organizational, not technological. The difference between AI leaders and laggards isn't access to better models or bigger budgets. It's how the business is organized, governed, and equipped to support AI at scale.
For CFOs evaluating AI budgets: ask how investments will translate to measurable outcomes, demand clear ownership and governance structures before deployment, and ensure workforce readiness is a parallel priority alongside technology rollout.
For CTOs and CIOs: treat governance as an enabler (real-time monitoring, adaptive controls, clear escalation paths) rather than a barrier, and don't deploy multi-agent systems until ownership and accountability are defined across team boundaries.
For business leaders across functions: the 11% who are winning with AI didn't get there by experimenting harder. They got there by redesigning how their organizations operate to support AI at scale.
The $186 million question is whether your organization will be part of the 8% who see ROI — or part of the 92% still waiting for results to show up.
Want to calculate your own AI ROI? Try our AI ROI Calculator — takes 60 seconds and shows projected savings, payback period, and 3-year ROI.
Continue Reading
AI Strategy & Implementation:
- Banks Are Finally Getting Serious About Agentic AI — But Most Will Fail — Why 80% of agentic AI pilots never reach production
- Enterprise AI Adoption Now Outpaces Internet Growth — Stanford's 2026 AI Index data on 70% enterprise adoption
- Anthropic Moves to Usage-Based Billing: What It Costs You — CFO math on Claude's new pricing model
Sources:
- KPMG Global AI Pulse: Q1 2026 — Survey of 2,110 C-suite leaders across 20 countries
- GovInfoSecurity: What Enterprise 'AI Leaders' Are Doing Right — Analysis of KPMG survey findings
Questions on your AI strategy or ROI measurement? Connect with me on LinkedIn, Twitter/X, or via the contact form — I read every message.
— Rajesh
