How to Leverage AI for Business – What Most Leaders Get Wrong

How to Leverage AI for Business - What Most Leaders Get Wrong

By Caroline Kennedy

The AI Leverage Gap - Why Adoption Does Not Equal Value

Here is a number that should trouble every business leader in Australia. According to McKinsey's 2025 State of AI report, 78% of organisations now use AI in at least one business function. 71% regularly deploy generative AI across marketing, product development, service operations, and IT.

Those are adoption numbers. They look impressive.

Now here is the number that actually matters: only 5.5% of those companies report real return on investment at the enterprise level.

Read that slowly.

78% adoption. 5.5% return. That is not a rounding error. And the gap between those two numbers is not a technology gap. It is a leadership gap.

Most businesses have treated AI the way they treated the internet in the early 2000s - as a tool to bolt onto existing processes. Automate this. Speed up that. Make the current way of working slightly more efficient. But bolting AI onto a broken process does not fix the process. It accelerates the dysfunction.

The organisations that are actually generating value - the 5.5% - have done something fundamentally different. They have not just adopted AI. They have redesigned their businesses around what AI makes possible. And that is a leadership decision, not a technology decision.

How to Leverage AI for Business - The Three Levels

In my work with CEOs and leadership teams, I have observed a clear pattern in how organisations relate to AI. I call it the AI Leverage Ladder, and understanding where your business sits on it is the first step toward moving up.

Level One - AI as Tool

This is where most businesses are. AI is being used by individuals within the organisation to do their existing work faster. Drafting content. Summarising documents. Generating code. Answering customer queries through chatbots.

At Level One, AI is a personal productivity enhancer. It makes individuals more efficient, but it does not change what the organisation does or how it makes decisions. The business model is unchanged. The workflows are unchanged. The competitive position is unchanged.

There is nothing wrong with Level One. It is a necessary starting point. But if your organisation has been at Level One for more than twelve months, you have a problem - because your competitors are not standing still.

Level Two - AI as System

At Level Two, AI moves from individual tool to organisational system. This is where leaders begin redesigning workflows - not just automating tasks within them, but questioning whether the workflow itself should exist in its current form.

McKinsey's data shows that only 21% of companies have redesigned workflows end-to-end around AI. Those that have reported 10-20% cost reductions in engineering, manufacturing, and IT, and revenue uplifts of over 10% in marketing and product development.

Level Two requires leadership because it means making structural changes. It means looking at a process your business has run for a decade and asking: if we were building this from scratch today, with AI as a given, would we design it the same way? The answer, almost always, is no.

Level Three - AI as Strategy

Level Three is where AI fundamentally reshapes how the business creates and captures value. This is not about doing the same things better. It is about doing different things entirely.

At Level Three, AI informs strategic decisions. It shapes which markets you enter, which products you develop, how you price, how you serve, and how you compete. Gartner predicts that at least 15% of day-to-day work decisions will be made autonomously through agentic AI by 2028. The organisations preparing for that now - building the governance, the data infrastructure, and the leadership capability to work alongside autonomous systems - are the ones that will define their industries in the next decade.

Level Three is where McKinsey's estimated $2.6 to $4.4 trillion in additional value from generative AI actually lives. Not in faster emails. Not in automated summaries. In transformed business models.

The question for every leader reading this is simple: which level is your business operating at? And more importantly, which level are you leading toward?

The Psychology of the Leverage Gap

There is a reason most leaders stay at Level One, and it is not laziness or ignorance. It is cognitive comfort.

Level One feels productive. You can see it working. Your team spends fewer hours on administrative tasks. Reports get drafted faster. Customer queries are resolved more quickly. The feedback loop is immediate and satisfying. It feels like progress.

Level Two and Level Three feel uncomfortable. They require you to question systems that are currently working. They require you to tell your team that the process they have spent years perfecting is about to change. They require you to admit that you do not yet know what the redesigned version looks like.

This is above-the-line versus below-the-line leadership in its purest form. Above-the-line leaders sit with the discomfort of not knowing and move forward anyway. They take ownership of the strategic question and hold themselves accountable for the answer. Below-the-line leaders stay where it feels safe, justify the status quo, and wait for the technology to "mature" before committing.

I have watched this pattern play out in every major technology shift over the past 25 years. The leaders who wait for certainty before acting are always the leaders who act too late.

There is also a deeper psychological dynamic at play. Moving from Level One to Level Two requires a leader to relinquish control over processes they understand intimately. It requires trust - in the technology, in their team's capacity to adapt, and in their own ability to lead through ambiguity. For leaders who have built their identity around having the answers, that is a profoundly destabilising ask.

But here is the thing. Having the answers is no longer the job. Asking the right questions is the job. And the right question now is not "how can AI help us do what we already do?" It is "what should we be doing differently because AI exists?"

What Australian Leaders Need to Understand Right Now

The Australian context matters here, and it is more nuanced than the global headlines suggest.

The Department of Industry, Science and Resources reported in 2025 that while large Australian enterprises have broadly embraced AI, only approximately one third of SMEs have adopted it meaningfully. By early 2026, aggregated data shows that 64% of Australian small and medium businesses report using AI regularly - up from 39% in mid-2024. That is significant acceleration.

But Acceleration Toward What?

The same Australian data reveals that 22% of SMEs say they are not even aware of how to utilise AI. A further 40% are not planning to use it yet. And among those who are using it, the vast majority are operating at Level One of the AI Leverage Ladder - individual productivity, not systemic transformation.

Meanwhile, CSIRO's Data61 estimates that digital technologies, including AI, could contribute approximately $315 billion to Australia's GDP by 2030. That is not a projection based on current trajectories. That is a projection based on what becomes possible if Australian businesses move beyond adoption and into genuine leverage.

There is also a talent dimension that compounds the problem. The Reserve Bank of Australia noted in late 2025 that many Australian firms report significant difficulty finding skilled workers - data engineers, AI specialists, and the strategic leaders who can bridge the gap between technical capability and business value. As more organisations compete for these skills, the constraint will only tighten. The businesses that will navigate this successfully are the ones whose leaders understand AI well enough to direct it - not code it, but shape its application and hold it accountable to business outcomes.

The gap between where Australian businesses are and where they could be is not a skills gap or a technology gap. It is a leadership gap. And closing it requires a fundamentally different conversation at the top of every organisation.

The 5 Leadership Shifts that Unlock AI Leverage

Understanding the AI Leverage Ladder is one thing. Climbing it is another. In my experience coaching CEOs through this transition, five specific leadership shifts separate the organisations creating real value from those still tinkering at the margins.

Shift One - From "What Can AI Do?" to "What Should We Stop Doing?"

The most powerful AI question is not about capability. It is about elimination. Every process, every report, every meeting, every approval chain in your business exists because someone, at some point, decided it was necessary. AI gives you the opportunity - and the obligation - to challenge those assumptions.

I have worked with leaders who discovered that 30% of their team's weekly output was being generated, reviewed, and filed without ever being read by anyone. AI did not fix that problem. AI made it visible. The leadership decision was to stop doing it entirely.

That is leverage. Not automating waste. Eliminating it.

Shift Two - From Delegating AI to Directing It

Too many CEOs have delegated AI strategy to their technology teams. This is the equivalent of delegating business strategy to your IT department in 2005. It misses the point entirely.

Harvard Business Review's research in early 2026 confirmed what I have been seeing in my coaching practice: AI high performers are three times more likely to have strong senior leadership engagement in AI initiatives. Not oversight. Engagement. The CEO is in the room, shaping the direction, asking the hard questions, and holding the organisation accountable for outcomes - not just adoption metrics.

If you have handed AI to your CTO and stepped back, you have outsourced the most consequential strategic decision of this decade.

Shift Three - From Measuring Activity to Measuring Outcomes

Here is where most AI strategies fall apart. Organisations track how many people are using AI, how many tools have been deployed, how many processes have been "AI-enabled." Those are activity metrics. They tell you nothing about value.

McKinsey found that AI high performers set outcome-based objectives tied to business KPIs. They do not ask "are we using AI?" They ask "has AI changed our margin, our speed to market, our customer retention, our decision quality?" The difference between those two questions is the difference between Level One and Level Three.

Shift Four - From Experimentation to Integration

Gartner predicts that 40% of enterprise applications will feature task-specific AI agents by the end of 2026, up from less than 5% in 2025. But here is the sobering counterpoint: Gartner also predicts that over 40% of agentic AI projects will be cancelled by the end of 2027, due to escalating costs, unclear business value, or inadequate risk controls.

The pattern is familiar. Enthusiastic experimentation without disciplined integration. Pilot projects that never graduate. Innovation theatre that impresses the board but never reaches the customer.

The leadership shift required is from "let's try AI" to "let's build AI into how we operate." That means governance. That means data quality. That means change management. It is not glamorous work. But it is the work that separates the 5.5% from the 78%.

Shift Five - From Individual Upskilling to Organisational Redesign

The instinct for most leaders is to train their people on AI tools. Send them on courses. Buy licences. Encourage adoption. And that is necessary but radically insufficient.

The real shift is organisational redesign. It is asking: given what AI can now do, how should this team be structured? What roles need to exist? What decisions should humans make, and what decisions should be automated? How does information flow? Where does judgement add value, and where does it add friction?

These are not technology questions. They are leadership design questions. And they require the same strategic rigour you would apply to a market entry or an acquisition.

Why Most AI Strategies Fail - A Leadership Diagnosis

If you are reading this and recognising that your organisation's AI strategy has not delivered what you hoped, you are not alone. The data confirms that the overwhelming majority of businesses are in the same position.

But the diagnosis is rarely technical. In my experience, failed AI strategies share three common leadership patterns.

The delegation trap. The CEO treats AI as a technology initiative and hands it to the tech team. The result is solutions looking for problems, rather than problems driving solutions. AI becomes a capability without a strategy.

The activity illusion. The organisation measures adoption - how many people, how many tools, how many use cases - and mistakes activity for progress. The dashboards look healthy. The P&L does not.

The comfort zone anchor. The leader is willing to use AI to improve existing operations but unwilling to question whether those operations should exist. This is the fixed mindset applied to technology strategy. It is the organisational equivalent of getting faster at running in the wrong direction.

Each of these patterns is a leadership problem with a leadership solution. And the solution, in every case, starts with the person at the top being willing to rethink not just what their business does, but how they lead it.

Common Questions About How to Leverage AI for Business

What Does it Mean to Leverage AI for Business?

Leveraging AI for business means going beyond individual tool adoption to systematically redesign how your organisation creates value. It involves integrating AI into workflows, decision-making, and strategy - not just using it to speed up existing tasks. The distinction between using AI and leveraging it is the difference between incremental efficiency and structural transformation.

How Do I Know if My Business is Leveraging AI or Just Using it?

Ask yourself one question: has AI changed a strategic decision in your business in the past six months? If the answer is no, you are using AI at the individual level but not leveraging it at the organisational level. True leverage shows up in your margins, your speed to market, and your competitive positioning - not just in your team's daily habits.

What is the Biggest Mistake Leaders Make with AI Strategy?

Delegating it entirely to the technology team. AI strategy is business strategy. When the CEO is not actively engaged in shaping direction, setting outcome-based objectives, and holding the organisation accountable for value creation, AI becomes a collection of disconnected tools rather than an integrated capability.

How are Australians Businesses Performing with AI Adoption?

Australian adoption is accelerating rapidly. By early 2026, 64% of Australian SMBs report using AI regularly, up from 39% in mid-2024. However, most remain at the individual productivity level. CSIRO estimates that digital technologies could contribute $315 billion to Australia's GDP by 2030, but realising that potential requires a shift from adoption to strategic leverage.

What is the AI Leverage Ladder?

The AI Leverage Ladder is a framework for understanding where your organisation sits in its AI maturity. Level One is AI as Tool - individual productivity gains. Level Two is AI as System - redesigned workflows and processes. Level Three is AI as Strategy - transformed business models and competitive positioning. Most organisations are at Level One. Value creation accelerates dramatically at Levels Two and Three.

What Industries are Benefitting Most from Leveraging AI?

Every industry benefits, but the nature of the leverage differs. Professional services and retail are leading in hyper-personalisation and administrative automation. Manufacturing and logistics are seeing 10 to 20% cost reductions through AI-redesigned operations. The key variable is not industry - it is leadership willingness to move beyond surface-level adoption.

How Does Caroline Kennedy Help Organisations Leverage AI for Business?

Caroline works with CEOs and leadership teams to close the gap between AI adoption and AI leverage. Through executive coaching and strategic advisory, she helps leaders move from Level One thinking to Level Three execution - redesigning how they lead, how their organisations operate, and how they create value in an AI-enabled economy. Her approach is grounded in over 25 years of leadership experience, including running organisations with revenue exceeding $250 million.

 

Sources:

McKinsey & Company. The State of AI in 2025: Agents, Innovation, and Transformation. (https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai). 2025.

Gartner. Gartner Predicts 40% of Enterprise Apps Will Feature Task-Specific AI Agents by 2026. (https://www.gartner.com/en/newsroom/press-releases/2025-08-26-gartner-predicts-40-percent-of-enterprise-apps-will-feature-task-specific-ai-agents-by-2026-up-from-less-than-5-percent-in-2025). August 2025.

Gartner. Gartner Predicts Over 40% of Agentic AI Projects Will Be Canceled by End of 2027. (https://www.gartner.com/en/newsroom/press-releases/2025-06-25-gartner-predicts-over-40-percent-of-agentic-ai-projects-will-be-canceled-by-end-of-2027). June 2025.

Department of Industry, Science and Resources. AI Adoption in Australian Businesses for 2025 Q1. (https://www.industry.gov.au/news/ai-adoption-australian-businesses-2025-q1). 2025.

CSIRO Data61. Australia's Artificial Intelligence Ecosystem: Growth and Opportunities. (https://www.industry.gov.au/sites/default/files/2025-06/australias-artificial-intelligence-ecosystem-growth-and-opportunities-june-2025.pdf). June 2025.

Harvard Business Review. Survey: How Executives Are Thinking About AI in 2026. (https://hbr.org/2026/01/hb-how-executives-are-thinking-about-ai-heading-into-2026). January 2026.

Reserve Bank of Australia. Technology Investment and AI: What Are Firms Telling Us? (https://www.rba.gov.au/publications/bulletin/2025/nov/technology-investment-and-ai-what-are-firms-telling-us.html). November 2025.

 

If your executive team is ready for that conversation, Caroline Kennedy works with leadership teams across Australia and NZ to facilitate bespoke executive strategy sessions on AI leadership and change management. These are not just keynotes. They are high-accountability working sessions designed specifically for the boardroom - moving executive teams from awareness to ownership in a single, structured session.

The session - Leading Through the Gravity of Change: AI Strategic Change Management - is available as a 2 to 2.5-hour bespoke keynote and facilitated executive workshop, tailored to your organisation's specific context, leadership priorities, and stage of AI adoption.

Caroline also delivers this content as a keynote for corporate conferences, leadership summits, and industry events across Australia, New Zealand, and internationally. Click here for more info on her AI Keynote.