The AI-enabled Enterprise: Optimising your operating model with AI

AI and its return on investment is dominating Executive and Boardrooms globally. But amid the buzz, it’s easy to lose sight of what really matters. It’s easy to lose sight of the learnings that we’ve had from many failed attempts at digital transformations. 

This post shares our emerging point of view on AI as it relates to our current operating context, the opportunities for how it can be used to improve strategy execution, and the patterns and risks to consider. 

Living in the age of AI 

Like past waves of technology innovation, from steam power to digital, AI is a technology that is already changing how we work. And it isn’t just about the tools and technology. 

The real impact of AI is coming as we shift how we operate, organise and behave to unlock its power.

The extent of change that AI brings is across 3 dimensions:

• Individuals – Everyone working within large organisations needs to unlearn and relearn how they do their work today to enhance their own individual productivity.

• Teams – Teams need to re-think how they operate to factor in AI as a team member and leverage AI to improve the processes, systems, and products/services that they are delivering.

• Enterprise – Enterprise needs to re-design how they do what they do to use AI to improve the strategy to execution system. 

  
  

Who will get the biggest advantage will be those individuals, teams, and organisations that can change first, and make it scalable and sustainable!

The operating context for many large organisations

Many large organisations are operating in increasingly complex environments that are changing even faster than before. We’ve gone from a VUCA world – volatile, uncertain, complex, and ambiguous – to BANI – brittle, anxious, non-linear, and incomprehensible.

This means our operating context is:

• Continuously shifting across markets and customer sentiment that we need to sense/respond and provide clarity of strategic direction within.

• Interconnected more than ever before and work is increasing in complexity as leaders need to navigate functional silos to deliver.

• Generating data faster and at larger volumes than we can process.

• Overwhelming for teams that are busy trying to navigate the various tasks, tools and competing priorities, and on the constant verge of burnout.

This leads to real cost - missed insights, slower adaptation, reduced throughput, and ultimately, value leakage.

This is where AI can be an enabler - to augment human thinking, not replace it. Used well, AI can help organisations reduce overload, stay aligned, and execute strategy with more flow and less friction.

From intent to impact: 5 foundational patterns for AI

“AI is not like any other software. It is not predictable, reliable and does not follow a set of rules. It does not have an operating manual.” - Ethan Mottick, author of Co-Intelligence 

Based on past experiences from digital transformation and the fast-paced, ever-changing, always-enhancing nature of AI, we believe there are 5 foundational patterns to getting this right.

1. Start with the Why and What

   
   

Pattern:

Start with clarity of the strategic intent to define why, the case for change, and how you will measure success. [See the metrics section of this blog for suggestions on how to frame up defining success.]

Anti-Patterns: 

The why for AI is ‘adopting the cool new tech’ and following the trend.

Success defined by cost reduction and FTE savings gained. This will erode the psychological safety that you require within your organisation to adopt and deliver using AI.

2. Set-up lean governance and controls

   
   

Patterns:

Define ethical guardrails, establish lean compliance practices, embed risk / security requirements into the work, and establish clear decision rights.

Delegated accountability is the only way to achieve the required pace of decision-making. This means that AI learning cycles need to happen weekly or more frequently, demanding a 10x acceleration in governance and control responsiveness.

Rapid AI adoption demands heightened risk and security awareness. This means addressing privacy risks from bad actors using GenAI for fraud arising from employees inputting sensitive company/customer data into LLMs. [1]

Anti-Pattern:

Traditional governance processes and structures will hamper the pace of innovation and experimentation e.g. setting up another Governance Board to police and enforce governance. Only 25% of leaders feel that their organisation is prepared to address the governance and risk issues related to GenAI adoption. [2]

3. Build AI capability at scale

   
   

Patterns:

Upskill teams, evolve roles, and enable people - at all levels - to leverage AI.

Shift towards being a continuous learning organisation where you upskill regularly, share learnings openly, and learn from others always.

To follow Spotify’s lead, set AI as a base expectation for everyone in the organisation, include it in performance reviews, and encourage learnings to be self-directed.

Anti-Patterns:

Constraining AI development to a central AI team or your IT function will not set you up to scale impact. This means investing in your own people, at all levels, to build AI skills. Only 9% of large organisations globally have upskilled 50% or more of their workforce. [3]

Outsourcing your AI delivery to a 3rd party delivery partner (aka consultant / system integrator) will not yield optimal outcomes as your own people are the best people to work out how to use AI. AI tech capability is not standing still. This means that investing in a one-time AI training course will not suffice. 

4. Uphold technical excellence

   
   

Patterns:

Invest in the right data, platforms and tools that integrate with the ecosystem. 

Coordinate investment and contracts for AI platforms across the enterprise to help to manage costs and license fees.

Adjust your risk appetite and procurement processes to be able to take advantage of new/emerging AI tech.

Anti-Patterns:

Silo decisions made on AI tools within your organisation leading to high license costs.

Buying and choosing AI like another ‘big enterprise software’ will limit your choices. AI products are new and emerging. Leaders will need to explore and be aware of various start-up AI products/platforms to get the best tools for the job.

5. Experimentation as the delivery default

   
   

Patterns:

Your delivery approach should be guided by experimentation with a “Think big, Start small, Learn fast” mindset.

Focus on outcomes not outputs and prioritise based on value. Compared to the digital era, AI requires teams to operate at unprecedented rates, compressing learning loops significantly from months to days or even hours. 

Balance will need to be struck between the innovation tax of trying things multiple times and strategic implementation.

Anti-Patterns:

Use of traditional project management delivery methods as they are not set-up for test / learn. 

These 5 foundational patterns - supported by the right mindset and behaviours - will help enable a higher return on investment in AI. It is important to note that we would advise an iterative approach to building out these 5 building blocks – a thin slice to establish the right foundations and then to refine based on learnings, rather than a big bang approach to set-up all foundations before getting started with AI experimentation.

Reimagining the Operating Model with AI

AI has the potential to touch every layer of how an organisation operates as its value lies not in isolated use cases, but in how it enhances coherence, clarity, and pace across the operating model.

We’ve identified 5 hypotheses for where AI can enhance how you interact, deliver and create value for your customers.

1. How we interact with customers

Hypothesis: What if every customer interaction could be more timely, more relevant, and more connected?

With AI’s ability to personalise at scale, analyse sentiment, and anticipate intent, there’s a chance to shift from reactive service to proactive engagement to build trust and loyalty through responsiveness.

2. How we define strategy and work (OKRs and alignment)

Hypothesis: What if AI could help keep strategy and execution in sync, not just during planning, but in real time?

By connecting objectives/OKRs to activity and surfacing where work is drifting, AI has the potential to improve alignment, make progress more visible, and support better decision-making at every level.

3. How we organise teams and deliver (Operating model and ways of working)

Hypothesis: Could AI give you a clearer view of how value flows and where it gets stuck?

With real-time insights into work patterns, dependencies, and inefficiencies, AI can offer a way to continuously refine team design, optimise resource allocation, and improve flow without needing to restructure.

4. How we create and deliver E2E value (Product, Delivery, Tech)

Hypothesis: Could AI help delivery teams move faster with greater confidence?

By predicting risks, automating checks, and simulating outcomes, AI could reduce delivery friction and increase responsiveness, especially in environments where speed and quality are both critical.

5. How we behave and lead (Culture and decision-making)

Hypothesis: What if you could spot early signs of cultural tension or team stress before they became visible in outcomes?

AI can surface behavioural signals and enable lightweight nudges creating opportunities to reinforce desired mindsets, support psychological safety, and strengthen leadership effectiveness through timely feedback.

Making it real - being clear on what you are optimising for when leveraging AI

The impact of AI should be measured by how it improves outcomes that matter, not through vanity metrics or activity counts. Value realisation for AI is still early, and organisations are still in ‘discovery mode’. It is the perfect time to set-up a good data baseline of measures for how you will monitor and measure AI success.

AI should contribute to a healthier system and delivery of better outcomes — one that delivers Better quality, realises more Value, delivers Sooner, operates Safer, and makes employees, customers, and communities Happier.

Beware the pitfalls: AI risks to actively manage

AI brings both opportunity and risk. It’s not plug-and-play. There are key risks to actively manage which include:

• Cost pressure: Focus on outcomes, not isolated ROI.

• Ethical bias: Address bias and build transparency.

• Privacy & security: Strengthen safeguards and compliance.

• Workforce impact: Prepare people through change readiness.

• Reliability: Keep humans in the loop for critical decisions.

• Vendor lock-in: Design with ecosystem flexibility in mind.

This isn’t about fear - it’s about responsibility. When used with care and clarity, AI can be a powerful catalyst for better work.

Summary – A call to action

You need to decide if you are going to be an organisation that simply uses AI or one that is enabled by AI. 

Being enabled by AI means putting humans at the centre, being clear on purpose, and using AI to enhance, not replace, our capacity to think, decide, and act.

The organisations that thrive in the Age of AI will be those that blend technical capability with human judgement, at scale, and those that adjust how they operate, organise and behave to leverage its powers.

__________________________________________

References

[1] Deloitte. (2024). State of Generative AI Report

[2] Deloitte. (2024). Deloitte Survey Report on realizing value while managing risks

[3] McKinsey. (2025). The State of AI Report