The 89% Problem: PwC's 2026 Operations Survey Reveals Why AI Investments Stall Before They Scale

Executive Summary

• 767 U.S. operations and supply chain leaders surveyed by PwC in early 2026. Respondents run companies with $100M+ annual revenue across eight industries — the same executives who sign off on ERP upgrades, warehouse automation, and AI pilots.
• 89% say their technology investments have not fully delivered expected results — yet 85% simultaneously claim they are ahead of competitors. The gap between self-assessment and outcomes is the defining feature of enterprise AI right now.
• Only 27% have fully embedded an AI strategy across business units. Only 37% are comfortable assigning AI agents to execute full end-to-end operational processes. The ambition-to-action gap is not a knowledge problem — it is a data and structure problem.
• 87% cite poor data quality as the primary reason digital initiatives fail to deliver value. Only 30% report significant improvement in data quality and reliability. Data readiness is the pre-condition most organizations skip.
• 4% meet all four success criteria simultaneously: AI embedded enterprise-wide, no significant barriers to scaling autonomous agents, horizontal operating structure, and technology investments delivering expected results. For every 25 companies that started this journey, one is actually winning.

The Self-Assessment Trap

The most revealing number in PwC’s survey is the gap between two responses that should not coexist: 85% of respondents say they are ahead of most competitors in digital transformation. 89% say their technology investments haven’t fully delivered expected results.

Both cannot be true at scale — if 85% are ahead of competitors, most companies by definition cannot be underperforming on outcomes. What this gap actually measures is a cognitive pattern well-documented in organizational psychology: leaders assess their intent and investment as positioning, not their outcomes. They bought the tools, hired the people, launched the pilots. That counts as “ahead” in their mental model. Whether it produced results is a separate question they are less comfortable answering.

This pattern has a practical consequence: it makes AI investment conversations inside companies much harder than they need to be. An executive who believes the organization is already ahead of competitors will not authorize the workflow redesign and data remediation that actually unlock value — those feel like admissions of failure, not forward motion.

The Three Bottlenecks Operations Leaders Actually Face

1. Data Quality Comes Before Agents

83% of respondents say AI agents and automation will accelerate the breakdown of traditional functional silos. Only 37% are comfortable assigning AI agents to end-to-end operational processes today. The gap between those two numbers — intent vs. comfort — sits almost entirely on top of a data problem.

87% say poor data quality has hampered their progress in achieving value for digital initiatives. Only 51% say their companies establish a clean, structured data foundation before scaling digital initiatives. Only 30% report significant improvement in data quality and reliability.

The implication is direct: organizations are launching AI agents on top of unreliable data and then attributing the failure to the agents. The actual failure mode is sequencing — data infrastructure is not a prerequisite in most AI roadmaps, but it should be.

This finding aligns with the broader corpus. Forrester’s 2026 enterprise AI research identifies data readiness as the single largest barrier to ROI. Deloitte’s survey (n=3,235) finds that companies in “deep transformation” — the 34% capturing outsized returns — address data quality before deploying AI at scale, not concurrently.

2. Strategy Is Declared, Not Embedded

27% have fully embedded an AI strategy across business units. 72% rank automating operations as a Top 3 AI investment focus. The gap between prioritization and embedding is not about budget — it is about organizational structure.

94% of companies with siloed or partially integrated operating structures expect to shift toward horizontal, networked models. Only 41% currently operate that way. The expectation exceeds the reality by more than 2x. Companies are planning to reorganize for AI rather than reorganizing.

This is the same pattern the PwC AI Performance Study (n=1,217, separate survey) identified: the top 20% of AI value capturers are twice as likely to have redesigned workflows, not just deployed tools. The Digital Trends in Operations data puts specific numbers on the structural gap that produces that outcome differential.

3. The 4% Threshold Is the Real Benchmark

Only 4% of respondents meet all four criteria simultaneously:

1. AI fully embedded enterprise-wide
2. No significant barriers to scaling autonomous agents
3. Collaborative, horizontal operating structure
4. Technology investments fully delivering expected results

This is the operational equivalent of the PwC AI Performance Study’s finding that the top 20% capture 74% of AI’s economic value. The path to the 4% runs through the same two requirements: data foundation first, structure redesign second. Organizations that treat these as downstream from AI deployment — something to fix after pilots succeed — are statistically unlikely to reach the 4%.

Key Data Points

Source credibility note: PwC is a management consulting firm with commercial interest in operations transformation engagements. This survey covers operations and supply chain leaders specifically — a more operationally grounded sample than the typical C-suite generalist survey. No independent verification of findings is available. Cross-reference against: Deloitte State of AI in the Enterprise 2026 (n=3,235, 34% deep transformation), PwC AI Performance Study 2026 (n=1,217, top 20% capture 74% of value), Forrester GenAI Enterprise Value 2026 (independent; consistent with data-readiness-first finding).

What This Means for Your Organization

The 89%/85% gap is not just a data point — it is the diagnostic. If your leadership team believes the organization is ahead of competitors while simultaneously acknowledging that technology investments haven’t delivered, the conversation you need to have is not about which AI tools to add. It is about sequencing.

The PwC data aligns with every credible enterprise AI study in one specific way: the companies that capture measurable value do data remediation and workflow redesign before scaling AI deployment, not after. The 4% threshold makes this concrete. Meeting all four criteria is not about budget — companies that spend more are not significantly more likely to reach it. It is about doing the unglamorous infrastructure work (data quality, organizational structure) that makes the tools work.

For COOs and supply chain leaders specifically, the 37% comfort rate with end-to-end AI agents is an honest number. It reflects where the actual production risk sits. Agents operating on unreliable data in siloed structures will surface errors faster than humans did — the visibility is higher, which means the accountability is higher. The organizations piloting agents successfully are almost always ones that cleaned the data environment first.

If the numbers in this briefing raised questions specific to your operations and supply chain context, the conversation is worth having — brandon@brandonsneider.com.

Sources

Corroborating sources:

• Deloitte State of AI in the Enterprise 2026 (n=3,235): 34% deep transformation vs. 66% efficiency-only — research/04-consulting-firms/deloitte-state-of-ai-enterprise-2026.md
• PwC AI Performance Study 2026 (n=1,217): top 20% capture 74% of AI economic value — research/04-consulting-firms/pwc-ai-performance-study-2026.md
• Forrester GenAI Enterprise Value 2026: data readiness as primary ROI barrier — research/05-analyst-firms/forrester-genai-enterprise-value-2026.md

Brandon Sneider | brandon@brandonsneider.com May 2026