AI and Business Improvement

 

AI and Business Improvement.

 

I, like many of you reading this, will be interested to varied degrees in how AI can help improve business improvement. In a world of change and challenge, where businesses want more for less from teams and budgets, and clients or customers expectations increase, companies are seeking to improve business processes, procedures and performance. Thus its reasonable to examine how AI can better enable business improvement.

At this point we know that the AI offering is a moving fast, and this article accepts that, and is prompted by a conversation last week with 5 c-suite colleagues and clients with various degrees of AI knowledge and interest, from health care, banking and tech.

Suffice it to say they are in part all confused by or in the confusion that AI is stirring up.

So here's some thoughts, boosted by some stats, from our chat for your consideration.

 

What works, what doesn’t and why Human + AI work better together.

One of the biggest themes in business today is that we’re 'drowning in data but starving for decisions', as one of my clients recently put it.

AI promises to change that and in many ways it already has. Yet for all the hype around AI automation and prediction, the real barriers are not technological from what I hear from clients. They are the human elements or barriers such as culture, skills, judgement and leadership. To improve business performance in a meaningful and sustainable way, organisations need to understand what AI can do, what it can do for them, where it falls short and how humans must be part of the equation.

 

The current state of AI adoption and outcomes

AI adoption is now widespread, though often staff are using their own choice of AI at work, and a recent survey found that 93% of firms use AI in some form, primarily in customer service, data forecasting and decision support systems. These implementations improve analytical speed, reduce error and increase the clarity of managerial discussions if not decisions. Yet organisations still face barriers including resistance to change, implementation costs and regulatory uncertainties, rather than purely technological limitations to change and improvement.

From a broader industry view, key 2025 statistics highlight both the potential and the gap in results:

• 65% of senior executives expect AI and predictive analytics to be key drivers of growth, with over half reporting significant productivity gains and faster ideation.

• Organisations using AI in decision making report a 50% increase in decision speed and 11% higher profitability than peers. (from Azumo)

• Only about 11% of companies say the majority of their AI initiatives have delivered tangible results, with most still in pilot stages. (TechRadar)

• In a broader global study, just 5% of companies are meaningfully benefiting from AI investments. Their success is linked to strategic leadership, strong data foundations and deeply integrated workflows. (Business Insider)

 

These numbers tell us that AI can deliver transformational impact, but very few organisations have cracked the code to make it work at scale and the bottleneck is rarely the technology.

 

Where AI can improve business

AI excels in areas where scale, speed and pattern detection are critical. These are domains where human cognitive limits are real and where AI’s computational power offers clear advantage. Here's some points for consideration

1. Data analysis and insights

AI systems can analyse vast datasets both structured and unstructured, far faster than humans. The value here is deeper faster insights, better forecasting and more informed strategic planning. Firms using AI-driven analytics report around 30% increases in operational efficiency and 25% boosts in predictive accuracy for inventory and resource planning. (MoldStud)

2. Automation of routine tasks

For repetitive, rule based work, it seems AI and intelligent automation deliver significant gains. AI-powered automation can reduce errors, speed throughput and should free humans from mundane tasks such as basic document processing, scheduling and data entry. This is where organisations report seeing quick efficiency wins.

3. Enhanced decision support

AI can augment decision making by highlighting trends, correlations and scenarios that humans would (might) miss. Up to 44% of C-suite leaders trust AI insights enough to override their own decisions and many organisations report competitive advantage due to faster, more data-driven decisions.

4. Customer experience and personalisation

Natural language processing (NLP) and recommendation algorithms power chatbots, virtual assistants and personalised offers - that we experience daily - in ways that improve responsiveness and customer engagement. AI can handle first line interactions, paving the way the for human touch intervention where nuance, emotion or more personalised service is involved.

5. Operational process optimisation

Experimental ‘AI + Big Data’ process models show dramatic improvements including 42% faster processing time, 28% better resource utilisation and 35% lower operating costs in tested enterprises, demonstrating how AI can reshape end-to-end process efficiency.

 

Where AI can’t improve business without Humans

Despite AI’s strength in analytics and automation, it has clear limitations (certainly at this point in time). These limitations define why AI will not replace humans and why hybrid performance is the realistic path forward.

1. Contextual judgement and (business) nuance

AI lacks true contextual awareness, it doesn’t understand history, culture, norms or organisational subtleties the way humans do. Decisions that require nuance, ethical judgement or sector specific insight still rely on human understanding. AI may tell you the 'what' but not always the appropriate 'why' or 'how'. ( from ImpressIT)

2. Creativity, empathy and emotional intelligence (EQ / Eq)

Tasks involving interpersonal sensitivity such as leadership, negotiation, morale building, conflict resolution and such like remain human domains, and remain areas that need attention and investment. Microsoft’s CEO recently argued that EQ is a 'workplace superpower' in the AI age, eg intelligence without empathy is incomplete. (Business Insider)

3. Ethics, bias and accountability

AI systems can replicate biases present in their training data, producing discriminatory or unfair outcomes if unchecked. Human oversight is essential to ensure fairness, ethical compliance and responsible use. Humans must validate, question and correct AI outputs.

4. Tacit knowledge and judgement

Polanyi’s Paradox is the idea that we know more than we can explicitly explain or share with others, like driving a car, and it highlights the gap between data patterns and human intuition. Tacit knowledge — the deep, instinctive understanding of how things actually work — cannot be encoded easily into a machine. AI can process explicit information but currently struggles with (human) tacit judgement.

5. Psychological and social impacts

Overreliance on AI risks eroding collaboration and human connection, with early studies pointing to issues like isolation and reduced teamwork when human input is minimised. Maintaining collaboration and human interaction is critical for workplace culture and mental health. (Financial Times)

 

The human side - skills and conditions for business improvement

If AI automation handles scale and routine then humans must focus on the uniquely human skills that AI cannot replicate. These include, for example:

1.Cognitive and strategic skills

• Critical thinking
• Systems thinking
• Creativity

These allow humans to interpret AI outputs and frame strategic action. (Workplace journal)

2. Emotional and social intelligence

• Empathy
• Leadership and influence
• Negotiation

These skills enable collaboration and trust building which are essential for implementation. (Business Insider)

3. Ethics and accountability

Understanding the societal and ethical consequences of business decisions ensures AI systems are used responsibly. Human governance - the final say - matters more now, than ever before.

4. AI literacy and collaboration skills

Understanding AI’s capabilities, limitations and how to interact with it (e.g. effective prompting, evaluation of outputs) becomes essential for all knowledge workers. Yet only about 34% of companies currently require AI training, seriously limiting ROI.

Upskilling modern workforces to collaborate with AI is not just technical, it’s cognitive, emotional and cultural. Leaders and teams need training, coaching and development programmes that go beyond coding to include interpretation, ethics, collaboration and decision making frameworks.

 

The optimal model of hybrid Human–AI.

Research consistently shows that human–AI augmentation outperforms either alone. When AI handles data processing and surface insights, humans can focus on interpretation, strategy and execution. In systems combining AI insights with human judgement and collaboration:

• ROI improves substantially. Strategic adopters achieve twice the ROI of basic AI users. This includes more efficient decision cycles and better operational clarity.

• Human–AI collaboration moderates performance gains organisations that integrate AI with human roles outperform those that rely on either human or machine alone. Human oversight strengthens AI’s contribution to innovation and competitive advantage. (arXiv)

What organisations need to put in place to unlock real business improvement, organisations need.....

 

1. Strong data foundations

AI is data-hungry anbd without quality, integrated data its outputs can be unreliable. Many firms still struggle with fragmented or inconsistent data.

2. Deliberate AI skills development

Training should extend beyond technical proficiency to include ethical judgement, interpretation, change and leadership in hybrid environments. (McKinsey)

3. Adaptive leadership and culture

Leaders must support experimentation, transparency and psychological safety to create spaces where humans can interact with AI, challenge outputs and collaborate without fear of retribution.

4. Clear governance and evaluation

Organisations need governance frameworks that define responsibility, accountability and performance metrics aligned with AI use, from ROI to ethical compliance and fairness.

5. Human-centred design

AI solutions need design thinking and human participation early in development otherwise tools will be underutilised or misaligned with real workflows. (McKinsey & Co)

 

In conclusion... humans still matter but AI changes the game

AI has real power to transform business performance with potential efficiency gains of up to 25% in productivity, 41% reduction in errors, faster decisions and competitive advantages. But these gains are only realised when organisations treat AI as augmentation, not replacement.

The future of business improvement is hybrid, humans and machines working together, each playing to their strengths. Humans bring judgement, context, empathy, creativity and ethics and AI brings scale, speed and pattern recognition. Together, they can make decisions faster, better and with more confidence thereby unlocking business value that neither could deliver alone.

To reap the full rewards of AI, business leaders must invest in tools and in people such as developing skills, reshaping culture, strengthening governance and creating environments where human–AI collaboration becomes the norm and not the exception.