Businesses love to talk about innovation. Many have innovation dashboards, labs, hackathons, or similar. Yet research of 4,595 businesses across the UK – based on data commissioned by the UK’s Department for Science, Innovation and Technology (DSIT) – found a very different picture underneath the rhetoric.
It explored the adoption of 20 critical ideas and technologies, including AI, looking at what is being adopted and the key barriers faced when doing so. Notably, UK businesses believe they are twice as innovative as they actually are. Almost half of businesses are yet to adopt any of the surveyed technologies, yet still rank themselves between seven and ten on a ten-point scale of innovativeness.
When it comes to adoption, over-optimism bias is both natural and understandable. Speaking about AI adoption and innovation can boost credibility, safeguard funding, shore up the share price, and preserve the leadership group – in the short-term, at least. The problem is that, in the long-term, optimism bias creates a dangerous gap between perceived progress and actual impact on AI innovation.
Outcomes, not inputs
Too often, businesses conflate a wider usage of technologies like AI with improved outcomes. This wouldn’t hold water in other areas. Take social media: by significantly increasing your friends and followers, you would certainly be enlarging your network, but you wouldn’t necessarily improve your meaningful relationships right away.
Similarly, many organisations benchmark their innovativeness based on inputs – R&D spend, labs, ideas generated – rather than outcomes. If you’re running pilots in generative AI, experimenting with computer vision, or partnering with a start-up, it’s easy to feel ‘ahead’. But there’s no guarantee that any of that work has reached your frontline operations or your customers.
Ultimately, innovation – the real reason behind any technology adoption – is about applying ideas, at scale, for impact. If an idea or technology doesn’t make a difference to people’s lives, it doesn’t matter. So, how can business leaders drive better AI outcomes?
Tie capability to a cause
For AI to be successfully adopted at scale, it needs to be rooted in fundamental need. If it doesn’t positively impact people’s lives, it doesn’t count. Delivering for impact means starting with demand, not supply.
For this reason, the use of technology needs to be driven by right-to-left thinking – starting with the desired outcome, not the technology itself. Rather than seeking to boost adoption, flip to a specific end goal such as cutting airport turnaround times by 15 percent, reducing speed-to-market for a brand by 10 percent, or uplifting fraud reduction for a bank. Focus on use cases where AI can measurably move mission metrics – not just where it looks impressive in a demo.
By tightening the frame of success, you should naturally tighten the frame of focus on the tools. AI is often referred to rather amorphously, but companies should focus more on which specific tools can best drive context-driven outcomes. Vastly different opportunities will arise from predictive analytics platforms, natural language processing tools, and computer vision systems, for instance.
Build the system – not the solution
While startups frequently come up with new ideas and technologies, the research shows it is medium and large organisations who are much more likely to adopt these – by a margin of five to one. In addition, organisations with overseas activity are more likely to be technology adopters, which underlines the importance of smaller firms connecting to larger enterprises.
AI adoption, as with other critical technologies, is really about the systems and structures around it. And this is two-fold. First, you need the supporting environment – from government, regulators, and the wider innovation ecosystem – for ideas to scale. These players can function as the connective tissue to help achieve greater market impact.
Second, you need the internal pathway: the governance, operating model, skills, and vendor ecosystem to build internal momentum. Progress is driven when you design from the outset for integration – ensuring AI systems plug into existing workflows, channels and decision‑rights rather than sitting as ‘shadow tools’. Crucially, AI adoption needs to be run more like a major change and transformation programme rather than as a short-term IT project or bolt-on – with accountable data and governance owners, and metrics that measure impact rather than track noise.
Invest in people, not platforms
Like all good quotes, the phrase ‘show me the incentive and I’ll show you the behaviour’ pre-dates the topic it’s best applied to. For it certainly rings true for many businesses when it comes to AI adoption.
Whether it’s in government, SMEs, or big industry, the difference between stalled and scaled AI programmes is almost always the same: do people understand what AI is for, how it changes their work, and what’s in it for them? Too often, AI budgets are geared around the tools themselves, with little regard for training, change and new skills. Yet when you give any emerging technology to teams, they’re not just receiving technology. They become active participants in a new working experience – and experiment – which is playing out in real-time, with knock-on repercussions to their roles.
For this reason, the best way for leaders to avoid the AI innovation mirage is to face the reality of the mirror, and ask: what does a good leader look like during an AI transformation? And how can I build confidence, respond to (natural) fears, demonstrate the right behaviours, and lead by example?
For example, organisations can now take quizzes to test their AI readiness in relation to their people. This helps leaders identify, for instance, where incentives are tilted towards incrementalism, such as when people are punished more for failed experiments than they are rewarded for successful ones. It also helps identify opportunities for role-specific enablement, as well as where frontline teams can be involved in co-design so that AI usage doesn’t feel imposed.
There is a mirage around innovation that often beguiles leaders. To achieve their goals, organisations must focus on purpose, not pilots. This means ruthlessly pursuing and investing in outcomes-driven adoption of the ideas that will move them forward. By tying capability to a cause, building systems (rather than purchasing ‘solutions), and investing in people rather than platforms, leaders can move from abstract ambition to tangible action – and from ideas to impact.
