AI Value Gap in Enterprise: BCG’s 2025 Blueprint for Future-Built Companies

Understanding the AI Value Gap: Why 60% of Companies Fall Behind

The AI value gap is no longer a theoretical concern — it is a measurable, accelerating divide that threatens the competitive position of most enterprises worldwide. Boston Consulting Group’s landmark 2025 report, The Widening AI Value Gap: Build for the Future, presents findings from a comprehensive survey of more than 1,250 senior executives and AI decision-makers across nine industries and over 25 sectors. The conclusions are sobering: despite more than $250 billion invested globally in AI during 2024 alone, a full 60% of organizations report minimal revenue and cost gains from their AI initiatives.

This research, which assesses AI maturity across 41 foundational capabilities spanning strategy, technology, people, innovation, and outcomes, reveals that the majority of companies simply do not have the proper capabilities for scaling AI in place. About one-third of these laggards admit they have made no meaningful progress at all. The question is no longer whether AI matters — it is whether your organization can close the AI value gap before the leaders pull irreversibly ahead.

What makes the AI value gap particularly dangerous is its compounding nature. Companies that moved early enjoy outsized benefits across financial and operational fronts, and they reinvest those proceeds into new capabilities, tools, and innovations. This creates a flywheel effect where success breeds further success, while laggards struggle to justify continued investment in the face of disappointing returns. As Nicolas de Bellefonds, BCG’s global leader of AI efforts, stated: “The companies that are capturing real value from AI aren’t just automating — they’re reshaping and reinventing how their businesses work. And they’re pulling away.” Understanding this dynamic is essential for any enterprise leader charting their AI transformation strategy.

Future-Built Companies: The 5% Driving Real AI Value at Scale

BCG’s maturity framework categorizes organizations into four tiers: stagnating (14%), emerging (46%), scaling (35%), and future-built (5%). The shift from 2024 to 2025 is notable — the scaling tier grew by 13 percentage points (from 22% to 35%), while the stagnating category shrank from 25% to 14%. Yet the future-built tier expanded by only a single percentage point, from 4% to 5%, underscoring just how difficult it is to reach the pinnacle of AI maturity.

Future-built companies are not simply spending more on technology. They are systematically building cutting-edge AI capabilities across functions and consistently generating substantial value. These organizations demonstrate near-universal C-suite engagement — nearly 100% report deeply engaged leadership teams, compared to just 8% of laggards. They are three times more likely to have appointed a chief AI officer and twice as likely to have a chief data officer in place.

The deployment velocity of future-built firms is equally striking. Some 62% of their AI initiatives are already deployed or scaling, compared to just 12% at lagging organizations. Their time-to-impact for AI workflows runs 9 to 12 months, versus 12 to 18 months for laggards. This acceleration is not accidental — it reflects deliberate investments in organizational design, talent development, and technology infrastructure that have been accumulating over multiple years. More than 60% of future-built firms rigorously track AI value, versus only 17% of stagnating companies, ensuring that every initiative is measured, optimized, and scaled based on evidence rather than intuition.

The Expanding AI Value Gap in Financial Performance

The financial consequences of the AI value gap are stark and growing. BCG’s analysis reveals that future-built companies achieve 1.7 times the revenue growth of laggards, 3.6 times the three-year total shareholder return (measured from June 2022 to May 2025), 2.7 times the return on invested capital, 1.6 times the EBIT margin, and 3.5 times the number of patents. These are not marginal differences — they represent fundamentally different trajectories for shareholder value creation.

Looking at the direct value from AI deployments, the numbers are equally revealing. In 2024, future-built companies reported a 6.2% revenue increase from AI initiatives, compared to just 1.2% for laggards — a 5.3x difference. On the cost side, future-built firms achieved 6.0% cost reduction versus 2.0% for laggards, a 3.0x gap. Projecting to 2028, future-built companies expect AI to drive a 14.2% revenue increase and 13.6% cost reduction, while laggards expect more modest gains of 6.8% and 9.6% respectively.

The investment differential tells its own story. Future-built companies spend 26% more on IT overall, allocate 64% more of their IT budget to AI, and invest 120% more in AI than laggards. The average global share of IT budget dedicated to AI stands at just 5%, suggesting enormous room for growth. One executive from a large multiformat retailer captured the strategic significance: “The investor community sees this, as we do, as a strategically important driver of value.” That company’s AI portfolio produced cost, margin, and revenue impacts of hundreds of millions of dollars over five years, adding more than 10% to EBITDA. For deeper analysis of how AI reshapes financial services, explore Oliver Wyman’s research on AI in financial services.

Agentic AI: The Next Frontier Widening the Value Gap

Perhaps the most striking finding in BCG’s 2025 report is the emergence of agentic AI as a powerful new force reshaping the competitive landscape. Barely discussed in 2024, agents already account for 17% of total AI value in 2025 and are projected to reach 29% by 2028 — nearly doubling their share in just three years. This shift is significant because agentic AI represents a fundamentally different paradigm: unlike traditional predictive or generative models, agentic systems learn, reason, and act autonomously to solve complex, multistep problems.

The current distribution of AI value by type shows predictive AI still leading at 45% (down from a dominant position in 2024), generative AI at 38%, and agentic AI at 17%. By 2028, the expected distribution shifts to 37% predictive, 35% generative, and 29% agentic. This convergence suggests that enterprises need capabilities across all three modalities, but the fastest-growing category — agentic AI — is precisely where the adoption gap is widest.

Future-built companies allocate 15% of their AI budgets specifically to agents, with a full third of these firms actively deploying agentic systems. Among scalers, just 12% use agents, and among laggards, the figure is effectively zero. Some 46% of all companies are experimenting with early pilots or deploying agents, with 16% already demonstrating tangible value. Meanwhile, 30% of companies report spending over 15% of their AI budget on agentic development.

The top functions prioritized for agentic AI usage are customer service (50%), customer journey (38%), R&D and innovation (37%), digital marketing (28%), and manufacturing (27%). Amanda Luther, BCG managing director and report co-author, emphasized: “Agentic AI isn’t a future concept — it’s already reshaping workflows and redefining roles. Companies should view it as the next step in scaling AI, not as the starting point.” Companies leveraging their broader ecosystem for agent development are three times as likely to experiment with or deploy agents. BCG identifies four sourcing options: standalone agentic solutions for narrow tasks, embedded solutions within CRM and ERP platforms, agent builder platforms for custom agents, and fully custom-built solutions for differentiating use cases. To understand the broader agentic AI transformation landscape, Bain’s complementary research provides additional perspectives.

Where AI Value Concentrates: Core Business Functions Over Support

One of the most actionable insights from BCG’s research is the concentration of AI value in core business functions rather than support operations. In 2025, 70% of potential AI value is concentrated in core functions — up from 62% in 2024 — with the remaining 30% in support functions. This shift challenges the common narrative that AI adoption should begin with back-office automation before moving to revenue-generating activities.

The breakdown of AI value across core functions is instructive: R&D and innovation leads at 15% of total potential value, followed by digital marketing (9%), manufacturing (9%), consumer journey (8%), sales (7%), digital supply chain (6%), maintenance (6%), pricing (5%), and core customer service (5%). On the support side, IT commands a disproportionate 13% share — up 6 percentage points from 2024 — reflecting the critical role of AI-powered development tools and infrastructure automation. Procurement, finance, customer support, and HR each account for 2-4%.

The data reveals that customer-centric and IT workflows together account for more than 50% of perceived AI benefits. This has important implications for resource allocation: companies that continue to spread AI investments thinly across all functions risk generating minimal impact anywhere. Future-built companies instead concentrate their efforts where value is highest, using rigorous prioritization frameworks to identify and scale the most impactful use cases first. For airlines and telecom companies, AI’s contribution to core function value reaches 80%, underscoring how industry context shapes optimal deployment strategies.

Five Strategies to Close the AI Value Gap

BCG’s research identifies five interlinked strategies that distinguish future-built companies from the rest. These strategies form a proven playbook that the other 95% of organizations can use to build AI maturity and achieve value at scale. Crucially, these are not sequential steps but interconnected capabilities that reinforce one another.

Strategy 1: Pursue a Multiyear Strategic AI Ambition

Future-built companies demonstrate 12 times more C-level executive engagement with AI than laggards. They treat AI not as a technology initiative but as a core business strategy with multi-year horizons, dedicated leadership, and board-level accountability. This top-down commitment ensures that AI investments are protected from short-term budget pressures and aligned with strategic priorities. Leadership engagement is the single strongest predictor of AI maturity — nearly 100% of future-built firms report deeply engaged C-suites, compared to just 8% of laggards.

Strategy 2: Reshape and Invent with Impact

Rather than simply automating existing processes, future-built companies fundamentally reshape how work gets done. They deploy five times more AI workflows than laggards, maintain 2.5 times more governance and value-measurement infrastructure, and achieve a 76% higher match between where AI is deployed and where it delivers impact. Nearly 90% of future-built and scaling companies expect most of their AI value to come from reshaping and inventing business processes, not from incremental optimization. The time difference matters: these companies achieve full deployment in 9 to 12 months, compared to 12 to 18 months for laggards.

Strategy 3: Adopt an AI-First Operating Model

The organizational design of AI implementation matters enormously. Future-built companies are five times more likely to leverage strategic workforce planning for AI and five times higher in maturity for implementing responsible AI guardrails. They embrace shared business-IT ownership, with 43% of future-built firms adopting equal responsibility models versus heavy IT-only ownership at lagging companies. Exclusive IT ownership is a strong indicator of stagnation — companies where IT alone drives AI implementation achieve significantly lower maturity.

AI-First Operating Models and Shared Governance

The governance structure of AI implementation emerges as a critical differentiator in BCG’s analysis. The data reveals a clear pattern: as companies mature in AI capability, they move away from both purely IT-driven and purely business-driven models toward shared ownership. Among future-built companies, 43% operate with equal business-IT responsibility, while just 7% maintain IT-only ownership. In contrast, 20% of stagnating companies rely on IT-only models, and 12% use business-only approaches.

This shared governance model addresses a fundamental tension in AI deployment: business leaders understand where value exists but often lack technical expertise, while IT teams possess implementation skills but may not fully grasp commercial priorities. Future-built companies resolve this tension by embedding shared ownership directly into their governance structures. More than 40% of future-built firms explicitly embed shared AI ownership into formal governance, compared to just 19% of stagnating organizations.

Responsible AI governance is another dimension where leaders pull ahead. Future-built companies are 4.6 times as likely to have fit-for-purpose guardrails in place and 2.6 times as likely to have rigorous tracking of AI value across the organization. This matters not only for risk management — 72% of companies already report unmanaged AI-security risks — but for maintaining trust with customers, regulators, and employees. A senior retail executive captured the sophisticated governance approach: “We concentrate in particular on senior sponsorship and ownership of AI benefits by the businesses, which creates the room to invest in foundational data. The AI capital allocation is managed centrally to make sure we are prioritizing the largest opportunities, but the amortization is carried by the business teams to ensure that they have skin in the game and own the impact.” This principle of effective AI governance has become a central concern for boards worldwide.

Talent and Upskilling: The Human Side of AI Transformation

The talent dimension of the AI value gap is perhaps the most underappreciated factor separating leaders from laggards. Future-built companies are six times more likely to dedicate structured time and programs for AI learning, plan three times more full-time equivalents for AI upskilling in the coming year, and are twice as likely to involve employees in shaping and adopting AI tools and workflows.

The scale of the talent investment is substantial. More than 50% of employees at future-built companies are expected to be upskilled in AI during 2025, compared to just 20% at lagging organizations. This three-to-one ratio in workforce development creates a compounding advantage: as more employees become AI-literate, they identify more use cases, generate more adoption, and accelerate the feedback loop between AI deployment and business value. Future-built firms also report that 50% more of their staff use AI in their daily work, creating a culture where AI augmentation is the norm rather than the exception.

The involvement of employees in reshaping workflows is a subtle but important distinction. Future-built companies involve their workforce twice as often in the process of redesigning how work gets done with AI. This participatory approach generates better outcomes because frontline employees understand the nuances of their work in ways that top-down implementations miss. It also reduces resistance to change by giving employees agency in their own transformation. Access to AI tools varies dramatically by sector: relatively mature industries provide more than 70% of employees with GenAI tool access, while less mature sectors reach fewer than 50%. Software companies plan to upskill 55% of their staff, compared to less than 15% at chemicals and machinery firms. According to McKinsey’s complementary research on AI talent, this skills gap is among the top barriers to enterprise AI adoption globally.

Technology Architecture and Data Foundations for AI Scale

The technology and data infrastructure underpinning AI initiatives is another area where future-built companies have built decisive advantages. BCG’s analysis shows these leaders maintain three times more standard templates and enterprise-wide data models, three times more central AI data policies defined and monitored by a central team, and three times more centralized AI platforms in operation to enable scale and adoption.

Data architecture, in particular, is a critical enabler. More than 50% of future-built firms operate on a single enterprise-wide data model, compared to approximately 4% of stagnating peers. This twelve-fold difference in data maturity directly impacts the speed and quality of AI deployments. Centralized data models reduce the friction of integrating new AI use cases, improve model accuracy through consistent training data, and enable cross-functional insights that siloed data architectures cannot support.

The technology sourcing landscape is more diverse than many assume. Only 11% of firms rely primarily on in-house AI development, and just 4% depend on a single end-to-end vendor for their full AI stack. Most companies adopt a hybrid approach, combining internal capabilities with external platforms and services. More than half of future-built companies maintain centralized repositories of reusable models and prompts, enabling faster deployment of new use cases while maintaining quality standards. This ecosystem-oriented approach extends to agentic AI as well: companies leveraging their broader ecosystem are three times as likely to experiment with or deploy agents. The Stanford AI Index corroborates these trends, documenting the rapid evolution of enterprise AI infrastructure globally.

Industry-Specific AI Maturity: Leaders and Laggards Across Sectors

BCG’s 2025 data reveals significant variation in AI maturity across industries, with important implications for competitive dynamics within each sector. Software companies lead the maturity rankings with an impressive 13-point year-over-year improvement on BCG’s index, followed by telecommunications (+11 points) and payments and fintech (+7 points). At the opposite end, fashion and luxury, chemicals, and real estate and construction remain in the early stages of AI adoption.

The industry-level data shows specific workflows creating scaled value across sectors. In insurance, claims validation and fraud detection workflows have achieved 50% deployment with 25% current impact, while underwriting optimization reaches 39% deployment. Energy companies lead with AI-powered infrastructure monitoring at 45% deployment. In consumer goods, demand forecasting and inventory optimization is at 23% deployment, while industrial goods companies are deploying AI-powered robotics at 29% deployment. Technology, media, and telecom firms see 20% deployment in product ideation workflows and 16% in self-service portals.

Regional patterns add another layer of nuance. North America leads in agentic AI experimentation at 51%, followed by Asia-Pacific at 45% and Europe at 41%. However, Asia-Pacific allocates the highest share of IT budget to AI at 5.2%, compared to Europe’s 4.6% and North America’s 4.4%. Asia-Pacific also leads in expected 2028 revenue impact from AI at 10%, versus 8% for North America and 7% for Europe. These regional variations suggest that the AI value gap is not only a company-level phenomenon but also a geographic one, with implications for global competitiveness and economic policy.

Real-world case studies illustrate the scale of opportunity. A global beauty products company deployed an industry-first virtual beauty assistant across more than 20 markets covering 8 brands, expecting $100 million in incremental revenue and doubling the ROI of traditional e-commerce pathways. A leading electronics manufacturer integrated GenAI across more than 200 factories, supporting 80% automation in complex operational workflows and modeling over $300 million in EBIT impact. As Michael Grebe, BCG managing director and report co-author, concluded: “For the majority of firms, catching up will require more than investment — it will take reinvention. The good news is that the playbook followed by future-built companies is clearly delineated and available to all.” For more on how AI foundations are developing globally, the World Bank’s complementary analysis offers valuable context.