Enterprise AI has moved beyond the question of whether organisations should experiment with it. Most already are. The more important question is whether those experiments are changing the performance of the business in a measurable, scalable and sustainable way. That is where the evidence becomes a little more uncertain. Are organisations getting the return on investment that AI promises?
According to McKinsey’s 2025 State of AI research, 88% of business leaders reported regular AI use in at least one business function, up from 78% the previous year. Yet McKinsey also identified only around 6% of organisations as AI “high performers”, defined as those attributing 5% or more of EBIT to AI. Fewer than 40% of companies reported any enterprise-level EBIT impact from AI at all, and most of those reported impact below 5%.
That gap between usage and value is the real AI story. It is not that AI is failing to capture organisational attention. It clearly has. The problem is that AI activity is being mistaken for AI progress. Licences are being purchased, pilots are being launched, innovation teams are being mobilised and employees are being encouraged to “use AI more.” But those things do not automatically create business value. They prove adoption intent, not transformation maturity.
The mistake many organisations are making is treating AI as a technology deployment rather than an operating model shift. They are trying to graft AI onto legacy workflows, fragmented systems, unclear decision rights, weak governance and data estates that were never designed for AI-enabled work. The result is often an impressive proof of concept that cannot survive contact with enterprise reality.
This is why AI readiness is so important. AI does not simply automate work; it exposes the quality of the organisation beneath it. If the organisation has poor data, AI exposes it. If workflows are fragmented, AI exposes them. If accountabilities are unclear, AI exposes them. If governance is underdeveloped, AI exposes it. If leadership alignment is weak, AI exposes it. AI can amplify capability, but it can also amplify confusion.
The problem is now becoming familiar. A business identifies a promising use case, runs a pilot, demonstrates potential and generates executive interest. Then the initiative hits the hard edge of scale: inconsistent data, security concerns, integration complexity, unclear ownership, operational resistance, lack of adoption planning, and no credible mechanism for measuring value. What looked like a technology opportunity becomes a transformation problem.
IBM-linked research has highlighted the same concern. Over the last three years, CEOs reported that only 25% of AI initiatives delivered the ROI they expected, while only 18% of companies achieved ROI above the cost of capital. That is a significant warning for boards and leadership teams. AI investment is increasing, but the operating discipline required to convert that investment into value is not keeping pace.
The opportunity, however, remains substantial. Boston Consulting Group’s 2025 research found that only 5% of more than 1,250 global companies are achieving measurable value from AI, while 60% have realised little to no benefit. BCG’s analysis points to a clear pattern among the successful minority: leadership engagement, long-term AI planning, reimagined workflows, workforce upskilling, and strong technology and data foundations.
That distinction is critical. The winners are not simply asking where they can insert AI into existing tasks. They are asking how work should change now that AI exists. That moves the conversation away from isolated automation and towards workflow redesign, decision quality, organisational capability, and measurable value creation.
This is where AI becomes a transformation question. The real value does not come from making old work slightly faster. It comes from redesigning the way work flows through the organisation. AI can reduce manual effort, accelerate analysis, support knowledge work, strengthen customer operations, improve software delivery and enable new forms of productivity. But those outcomes depend on whether the organisation has created the conditions for AI to be trusted, adopted and governed.
A practical AI readiness assessment therefore needs to look beyond tools and use cases. It should test whether the organisation is ready across multiple dimensions:
The maturity curve that emerges from this assessment is more useful than a simple “ready or not ready” judgement. At the lowest level, organisations are ad hoc. Individuals use AI informally, often with enthusiasm but without consistency, governance or a clear connection to business outcomes. At the next level, organisations are experimental. Pilots are launched, use cases are explored and early learning begins, but value remains localised and scaling remains uncertain.
The emerging stage is where structure starts to appear. Leaders recognise the need for governance, use-case prioritisation, data readiness and change planning. However, AI is still not fully embedded into operating models or value streams. At the operational stage, AI becomes connected to measurable outcomes, redesigned workflows, active governance and managed adoption. Finally, at the transformational stage, AI becomes part of how the organisation thinks, decides, serves customers and creates value. Human judgement is not replaced; it is amplified through better systems, better evidence and better ways of working.
Most organisations are somewhere between experimental and emerging maturity. That is not a failure. It is a realistic starting point. The problem comes when organisations overestimate their maturity because they measure activity rather than readiness or outcomes. A long list of pilots, licences and tools may create confidence, but it does not prove that the organisation is capable of scaling AI safely or profitably.
Leadership is decisive here. AI transformation cuts across strategy, finance, operations, technology, people, risk, compliance, data and customer experience. It cannot be delegated entirely to IT, an innovation lab or a group of enthusiastic early adopters. Gartner has predicted that by 2026, 75% of organisations will have operationalised AI, and that by 2027, 75% of CDAOs who fail to demonstrate AI’s positive impact will be reassigned or removed from the C-suite.
That prediction is telling. AI value is becoming a leadership accountability, not a technical experiment. Senior teams need to understand not only what AI can do, but what the organisation must become to use it well. That requires a more honest conversation about readiness, value, risk and change.
The strongest signal that AI adoption is a transformation problem is coming from the AI companies themselves. If enterprise value could be unlocked simply by making better models available or deploying tools, frontier AI companies would focus almost entirely on product, platform access and model performance. Instead, they are moving closer to consulting, deployment and organisational change.
OpenAI has created the OpenAI Deployment Company, backed by more than $4 billion in initial investment, to accelerate corporate AI adoption. Reuters reported that this includes the acquisition of Tomoro, an AI consulting firm founded in partnership with OpenAI, bringing approximately 150 AI engineers and deployment specialists into the new venture. Its focus is to embed specialists inside organisations to identify high-impact AI deployment opportunities and support implementation at scale.
Anthropic is taking a different but related route through enterprise consulting partnerships. Business Insider reported that Anthropic has significantly expanded its partnership with PwC to accelerate Claude adoption across corporate America, including plans for PwC to train 30,000 US employees in Claude Code and then scale the initiative across PwC’s global workforce of 364,000 people. The partnership also includes a joint Centre of Excellence focused on integrating AI into core business operations, including agentic engineering tools, dealmaking processes and operating-model reinvention.
This is important because it validates my central point: the hard part of AI is not just access to capability. The hard part is changing how organisations work. The world’s leading AI companies are recognising that models need deployment capability, advisory support, workflow redesign and change enablement if they are to translate into enterprise value.
For the largest enterprises, this gap will increasingly be served by the major consultancies, large-scale deployment companies and global systems integrators. They have the capacity, brand reach and enterprise relationships to support global banks, pharmaceutical companies, telecoms firms and multinational industrials. That makes sense at the top end of town.
But it leaves a different challenge for the mid-market.
Mid-sized growth companies, private-equity-backed businesses, and buy-and-build platforms face the same AI readiness problem, but without the same internal capacity, change infrastructure or consulting budgets as large enterprises. They still need to understand where AI can create value, whether their data and workflows are ready, how adoption should be managed, what governance is required, and how AI investment links to EBITDA, productivity, customer experience or operating leverage. What they often do not need is a huge transformation consultancy with its accompanying price tag. They need senior, pragmatic, independent support that can help them move quickly without creating unnecessary complexity.
This is where Arqvera can help.
Arqvera helps mid-market and PE-invested organisations move from AI experimentation to AI value realisation by treating AI as a transformation challenge, not a tool deployment. The objective is not to sell another AI tool. The objective is to help organisations become ready to extract measurable value from AI.
In practical terms, that means helping leadership teams clarify the business problem or opportunity before technology decisions are made; map AI use cases to measurable outcomes; assess data, governance, workflow and change readiness; identify where work needs to be redesigned; align stakeholders around value and risk; and create a practical roadmap that moves the organisation from experimentation to operational maturity.
This is critical particularly for PE-backed companies because AI adoption is now becoming part of the value-creation agenda. Used well, it can improve productivity, reduce operational friction, accelerate decision-making and create scalable capability without simply adding headcount. Used badly, it becomes another fragmented technology spend that distracts management, increases risk and fails to show up in measurable performance.
Arqvera’s point of view is that AI should amplify human potential, not bypass human judgement. That requires more than enthusiasm. It requires clarity, candour, consistency, care and continuity. It requires a structured approach to readiness, adoption, governance and value realisation using a framework like our AI.ccelerate. It also requires a willingness to ask harder questions before investment accelerates.
Before organisations invest further in tools, pilots or large-scale AI programmes, they need to understand where they are today. Not through optimism. Not through vendor promises. Through a clear-eyed assessment of the conditions that determine whether AI can create value and what the foundational and acceleration levers are to realise value.
That is the purpose of Arqvera’s AI Readiness Self-Assessment. It gives leaders a practical way to understand where their organisation sits on the AI maturity curve, where the biggest gaps are, and what needs attention before AI can move from experimentation to measurable value.
Take the assessment here: https://www.arqvera.com/ai-readiness-self-assessment
AI will not create enterprise value by itself. Ready organisations will.
Is an AI and technology transformation consultancy and advisory.
We help organisations shape business cases, projects, deliver excellence, and realise change and outcomes that stick. We support organisations before, during, and after projects with an end-to-end service where our domain specialization comes to life.
Before (Inception): We work with you to clearly define the idea, vision, strategy, and business case for change, as well as help select the right partners, and establish governance
During (Execution): We help deliver project and change objectives while keeping implementation under control through structured governance and assurance to realise intended outcomes.
After (Value Realisation): We ensure outcomes deliver measurable value and embed continuous improvement from successes and learnings.
Arqvera is led by industry veterans in the UK and USA with 100+ years of technology delivery intelligence across global consulting, digital transformation, and mission-critical projects and programmes.