We surveyed 501 technology leaders across healthcare, media, travel, and retail. The findings confirm what many of us already suspected: AI adoption is no longer the hard part. Execution is.
The gap between commitment and results is widening. Organizations have leadership buy-in, approved budgets, and active AI initiatives underway. The mandate is real, the momentum is real, but the outcomes? For most, they remain elusive.
We partnered with Redpoint Insights to field this research between April and May 2026. The goal was simple: understand what actually separates the companies scaling AI from the ones stuck after their first pilots.
The Problem Everyone Assumed Was Budget
When boards and leadership teams diagnose what is blocking AI progress, the conversation almost always starts with funding and talent. Do we have enough budget? Do we have the right people?
The data says those instincts are wrong.
Budget was the least-cited barrier in the entire study. Just 17% of respondents pointed to it. Skill gaps came in only slightly higher at 20%. These are real challenges for some organizations, sure. But they are not what is keeping the majority from getting AI out of pilot mode.
The top barrier? 42% of organizations say they cannot move AI initiatives from pilot to production. That is an operational problem, not a financial one. Another 37% struggle to measure whether their AI work is actually producing business results. And 33% cannot identify or prioritize the right use cases in the first place.
The pattern is clear. The hardest challenges emerge after implementation begins.
76% Cannot Prove AI Is Working
This might be the most important number in the entire study.
Three-quarters of organizations in our research cannot clearly quantify the business value or ROI of their AI initiatives. They believe AI is creating value, many are confident it is, but they lack the data to demonstrate it.
Without evidence of impact, teams cannot decide what to scale. Budget decisions become harder to defend. Investment spreads across too many initiatives because nobody can see where value is actually being created. Over time, AI efforts stall under their own ambiguity.
The organizations that break this pattern look fundamentally different from the ones that do not. The full report examines exactly how, and the measurement gap is one of the strongest predictive findings in the data.
Three Stages of AI Progress
We grouped the 501 respondents into three maturity tiers based on how far they have progressed, what outcomes their initiatives have produced, and whether they can measure business impact.
Builders (22%) are still exploring or piloting. They have limited integration and no visibility into ROI.
Executors (56%) are actively implementing. They have real projects in production and are navigating the difficult transition from pilot to scale.
Leaders (22%) have AI operating at a meaningful scale. Their production systems are running consistently and delivering measurable returns.
The divide between Executors and Leaders is where the most actionable insights live. Both groups have moved past experimentation, both have invested real resources. What separates them is a set of operational practices and organizational disciplines that the full report breaks down in detail.
44% Are Stuck in the Execution Gap
We also segmented organizations by where their AI efforts are getting stuck. The largest group, at 44%, falls into what we call the Execution Gap. These organizations have already delivered AI initiatives. The challenge is scaling and sustaining them.
Another 38% sit in the Mandate Gap. They have leadership support and budget in place, but the investment has not yet translated into meaningful outcomes.
The Execution Gap segment is the most strategically important group in the study. They are not lacking ambition or funding. They need the operating discipline to turn early progress into repeatable, measurable business capabilities.
What Leaders Actually Do Differently
The full report identifies five specific operational practices that distinguish the organizations scaling AI from everyone else. We will not spoil all of them here, but a few findings are worth previewing.
Leaders are far more likely to have defined metrics for measuring AI impact before implementation begins. That sounds obvious, but most organizations skip it. The correlation between measurement capability and scaling success is one of the strongest findings in the entire dataset.
Leaders also bring in external expertise earlier and more deliberately than less mature organizations. 66% of Leaders are very likely to engage an external partner in the next six months. The companies furthest along in AI execution are the ones most actively seeking specialized support. That finding challenges the assumption that needing help signals weakness. Actually, the data says the opposite.
The remaining practices, and the data behind them, are in the full report.
Industry Patterns Worth Watching
The core findings hold across all four industries we studied. Organizations are investing, struggling to scale, and looking for help. But the friction shows up differently depending on the sector.
Travel & Hospitality shows the strongest demand for external partners. Healthcare faces infrastructure and compliance constraints that slow execution despite having the strongest formal AI mandate. Retail & Ecommerce is the most stalled industry in the study, with just 36% having successfully scaled AI. Media & Publishing is furthest along in adoption but faces a different kind of uncertainty about how AI reshapes the way content is created, distributed, and monetized.
Each industry section in the full report includes specific data, respondent quotes, and a case study showing how the execution gap plays out in practice.
AI Is Changing Teams, Not Eliminating Them
One more finding worth flagging. The workforce data in this research does not support the simple automation narrative. More organizations expect headcount growth (45%) than reduction (38%). The real shift is in composition. 42% expect increased demand for senior engineers, architects, and oversight roles. 41% expect AI-specific responsibilities to be added to existing positions.
As AI takes over repeatable work, the work that remains becomes more complex. Organizations need people who can design systems, evaluate tradeoffs, govern implementations, and make the judgment calls AI cannot make on its own. The full report explores how this talent shift connects to the partner and execution findings across the study.
Read the Full Report
The AI Execution Gap report is based on a quantitative survey of 501 technology leaders, directors through C-Suite, across Healthcare & Life Sciences, Media & Publishing, Travel & Hospitality, and Retail & Ecommerce.
The full report includes detailed maturity-tier comparisons, industry-specific breakdowns, workforce data, partner engagement patterns, and the five operational practices that separate Leaders from everyone else.
