Most organisations encounter significant challenges in quantifying AI’s financial impact
AI is everywhere, or so it seems. But if you’re sitting on the board or running a billion-pound firm, the real question isn’t whether you’re using AI. It’s whether it’s delivering measurable returns. Right now, most companies can’t answer that confidently. Research from Cynozure’s 2026 State of the Industry Report paints a clear picture: only 15% of organisations can quantify the financial impact of their AI initiatives in actual pounds or dollars.
Highly capable teams are building advanced models and integrating generative and agentic AI. There’s clear demand and investment. However, there’s a gap between deploying AI and measuring its real contribution to the bottom line. This is a problem, and it’s commercial, not technical. Without strong financial metrics tied to AI-related work, many companies are flying blind. They might be automating processes or making data-driven decisions, but they’re not showing whether those efforts translate to revenue, cost savings, or margin expansion.
Executives need to shift how they treat AI, from experimental tools to business products with measurable performance. Run it like you would run a serious product line. Measure investment, tie it to expected outcomes, and track results. This shift makes it easier to justify future budget, secure board buy-in, and make smarter trade-offs.
If AI is to scale meaningfully across your business, it needs to earn its place in your P&L. Not just technically, but commercially.
While AI technology is pervasive, its benefits remain largely confined to operational and productivity improvements
AI adoption isn’t slowing down. From financial services to retail to consumer product firms, leaders are deploying AI models at scale. Generative AI models are widely used now; 52% of firms say they’re currently exploring or building on agentic AI, systems that can make and execute decisions with minimal human intervention.
Still, most of the impact is happening in one part of the business, operations. That’s fine, but it’s limited. The current focus revolves around automating repetitive tasks, reducing manual processing time, and boosting workforce efficiency. These are good, tactical wins. But they won’t move the market or change your company’s trajectory.
Strategic functions, like product innovation, go-to-market speed, or customer growth, are still underutilizing AI. You’re leaving value on the table if AI is only modernising internal tasks. To deliver outsized returns, AI has to extend into how your business creates value, not just how it runs.
If you’re already investing heavily in data platforms, models, and infrastructure, make sure the use cases are big enough. Routine automation delivers incremental gains. Strategic AI, tied to critical business levers, delivers exponential ones. Most aren’t there yet, but the opportunity is available to those who make the shift.
Ownership of AI strategy is fragmented, which impedes cohesive execution and clear accountability
For AI to drive impact, someone needs to own the strategy, completely. Not part-time. Not shared. Right now, that’s not the case in most companies. Cynozure’s 2026 State of the Industry Report shows 80% of organisations assign ownership of data strategy to a Chief Data Officer or Head of Data. That’s good. But only 28% extend that same clarity to AI strategy. Another 40% have split ownership across executives. And 17% have no clear AI lead at all.
This fragmentation slows everything down: clarity, prioritisation, accountability. If AI is everybody’s job, no one’s truly leading. Companies may still launch great technical experiments, but they won’t execute a unified roadmap. Without clear leadership, it’s tough to align AI to business goals, prioritise investments, or scale what works.
This isn’t a question of hierarchy. It’s about strategic execution. Someone has to own outcomes, own performance metrics, and make the hard calls on where AI delivers the most ROI. In some firms, that could mean expanding the scope of the Chief Data Officer to encompass AI. In others, it might require a dedicated head of AI or embedding leadership across specific verticals. Structure it how you like, but make it defined.
If you’re serious about AI, take the organisational ambiguity off the table. Appoint a capable lead with direct accountability. Make outcomes clear. Otherwise, execution will lag behind the technology you’ve already paid for.
Enhancing data culture and literacy is critical to unlocking the full potential of AI investments
AI tooling doesn’t create value unless people know how to use it, and trust the data it’s built on. This is where many companies run into friction. Despite the rise in AI adoption, internal data literacy is still lacking. Nearly half of senior leaders in Cynozure’s report cited “data culture and literacy” as their top priority for 2026. And they’re right. AI doesn’t work if people don’t understand or believe in the fundamentals feeding it.
When employees distrust the data or don’t know how to engage with AI-driven outputs, tools get sidelined. Projects stall. Momentum slows. You get partial adoption at best. And that means underperforming investments.
For executive teams, improving data culture is a long-term asset. Train teams to interpret data, question it constructively, and feel confident using data tools in day-to-day workflows. Make it standard, not a specialist activity. Promote transparency in how data is collected, processed, and turned into decisions. Trust follows clarity.
If you’ve already invested millions into AI infrastructure, but haven’t taken steps to upskill your workforce, you’re undermining the return. Data is a core business capability now. Raising organizational competence in this area sets the foundation for credible, repeatable wins with AI. And it puts your business in a better position for whatever comes next.
Organisations face distinct barriers in AI adoption influenced by size and existing technological infrastructure
Not all AI roadblocks are created equal. The challenges that a global corporation faces aren’t the same as those confronting a mid-sized firm. Cynozure’s 2026 State of the Industry Report makes that clear. Budget and resource constraints are the top blocker overall, cited by 25% of respondents. For smaller companies, this means AI remains out of reach or confined to limited, tactical pilots. Capacity is limited, and risk tolerance is understandably lower.
Larger organisations have different friction points. They’re not fighting to afford AI experiments, they’re blocked by what’s already in place. Twenty percent of leaders from large firms point to legacy systems as the key issue. Seventeen percent say it’s lack of executive or organisational buy-in. These firms have scale, but often not the agility, and without alignment from senior leadership, progress stalls.
For decision-makers, this highlights the importance of context. If your company is smaller, the challenge is funding and focus. If you’re running a larger operation, the challenge is often deeper, outdated infrastructure, internal resistance, or unclear executive support. Both are solvable, but the playbook changes depending on the size and structure of your organisation.
There’s no universal barrier to AI success, but there are universal consequences for delay. Whether it’s funding constraints or structural inertia, leadership teams need to solve for the specific blockers in front of them. Otherwise, strategic momentum is lost before scaled execution even begins.
Data products are emerging as pivotal tools to translate AI efforts into measurable business performance
AI only creates real business value when it’s tied to decision-making. If it doesn’t move the numbers, it doesn’t matter. One of the most practical developments right now is the emergence of data products, modular, outcome-oriented applications of AI and data designed for specific use cases. These are gaining traction because they translate technical work into something the business can use, measure, and scale.
In Cynozure’s survey, over 70% of leaders said they expect data products to deliver the most value in operational autonomy and performance. They’re not just about automating tasks or running dashboards. They’re about building reusable components that drive consistency, speed, and informed decision-making across the business.
By treating data and AI initiatives as product portfolios backed by performance expectations, businesses can track what works and what doesn’t. That makes it easier to validate ROI, justify future investment, and communicate meaningful progress at the board level. It also shifts AI out of the lab and into the business, where the results actually show up.
Tim Connold, Chief Client Officer at Cynozure, put it clearly: “Data products, and increasingly decision products, are how leaders are turning strategy into reality.” When organisations frame their AI investments around decisions, what to change, where to invest, how to allocate, they gain traction with boards, investors, and operational teams alike.
If your firm is already building models and pipelines, start thinking in terms of products. Not internal tools, not proof-of-concept tech, products with defined owners, usage expectations, and business performance targets. That’s how AI starts contributing to real business growth.
Emphasis on measurement discipline and governance maturity will define the next phase of AI adoption
We’ve moved past the phase where companies ask, “Can we use AI?” The answer is already yes. The real question now is, “Can we prove that it works, commercially?” That’s the next challenge. And most firms aren’t set up for it yet. To get there, organisations need stronger measurement frameworks and clearer governance structures. Without these, AI investment becomes hard to justify and even harder to scale.
As companies shift from experimentation to enterprise-level deployment, leadership expectations are also changing. Boards want proof, not just pilots. That means tracking output not only in technical terms, but in financial results: revenue impact, cost reduction, margin gain, or improved capital efficiency. Treating AI like a strategic asset, not a side project, is how real progress happens.
This requires mature governance. Someone has to account for how AI initiatives are scoped, budgeted, prioritised, and evaluated. Without that discipline, projects drift, and confidence erodes. Measurement frameworks should cover both performance and commercial outcomes. You don’t need dozens of metrics, you need the right ones, tied to your business model and strategic goals.
Jason Foster, Founder and CEO of Cynozure, said it plainly: “AI has propelled data into the boardroom. The challenge now is not whether organisations can use data and AI, but whether it is making a meaningful difference to the P&L.” He’s right. The organisations that lead in 2026 and beyond will be those that manage AI investments like they manage any other valuable asset: with clear outputs, strong accountability, and results tied directly to financial performance.
This isn’t just a reporting exercise. It’s about building a culture where AI earns its place, on the balance sheet, in decision-making, and in long-term strategic planning. If your AI programme isn’t linked to business outcomes you can measure, now’s the time to tighten it up. That’s where the next advantage will come from.
Final thoughts
AI isn’t a technical question anymore, it’s a leadership one. The infrastructure is there. The talent exists. The models work. What separates the frontrunners from the rest now is execution: who can translate AI investments into real, measurable business value.
That means treating AI not as a set of experiments, but as a portfolio of products aligned to outcomes. It means assigning ownership that’s accountable for performance, not just delivery. It means investing in culture and skills so your teams trust the data and understand how to work with it. And it means building the measurement systems that let you track return, not in abstract models, but in actual results.
You can’t scale what you can’t measure. And you can’t adapt fast if no one owns the roadmap. If those pieces are missing, expect diminishing returns.
But if you lead with discipline, act on data, and commit to commercial outcomes, AI won’t just be a capability. It’ll be a competitive advantage.
