What is an Operating Model
Source: Chandresh Uike via Pexals
summary || The Architecture of Execution: Understanding Your Operating Model
An operating model is the invisible system that turns strategy into results. It is the way decisions, resources, and information flow through an organisation to create value. Most models were built for a slower, more predictable world and are now breaking under the speed of digital, hybrid work, and AI. This article explains how operating models are evolving from traditional hierarchies built for control, to adaptive, AI-enabled systems built for learning and autonomy, and shows how leaders can design fit-for-purpose models that move value faster, align strategy with execution, and keep organisations ready for what’s next.
-
verb
gerund or present participle: operating
(of a person) control the functioning of (a machine, process, or system).
"the Prime Minister operates a system of divide and rule"
o (of a machine, process, or system) function in a specified manner.
"market forces were allowed to operate freely"
o be in effect.
"there is a powerful law which operates in politics"
o manage (a business).
"many foreign companies operate factories in the United States"
o (of an organization) be managed in a specified way or from a specified place.
"neither company had operated within the terms of its constitution"
o (of an armed force) conduct military activities in a specified area.
"the mountain bases from which the guerrillas were operating"
perform a surgical operation.
"surgeons operated on his jaw yesterday morning"
-
noun
a three-dimensional representation of a person or thing or of a proposed structure, typically on a smaller scale than the original.
"a model of St Paul's Cathedral"
a thing used as an example to follow or imitate.
"the project became a model for other schemes"
verb
fashion or shape (a three-dimensional figure or object) in a malleable material such as clay or wax.
"use the icing to model a house"
use (a system, procedure, etc.) as an example to follow or imitate.
"the research method will be modelled on previous work"
20min read
The Invisible System Running Your Business
Every CEO can describe their strategy in three slides. Most can sketch their org chart on a whiteboard. But ask them to explain their operating model, the system that actually turns strategy into results, and you'll get silence, or worse, a description of last year's restructure.
That gap matters. Because while strategy sets direction and structure provides the scaffolding, your operating model is the circulatory system of your business. It's how decisions, resources, and information pulse through your organisation to create value or don't.
Think of it this way: your org chart is a snapshot. It shows reporting relationships frozen in time. Your operating model is a film. It reveals how work actually flows across boundaries, how priorities shift when reality changes, and how fast your organisation can sense and respond.
An operating model answers the questions your org chart can't:
When two product teams both want the same engineering resource, who decides? How fast?
When customer feedback contradicts the roadmap, which meeting picks up the signal?
These aren't process questions. They're system questions. And most organisations have never designed for them. Instead, they've inherited a patchwork: legacy structures from the merger, governance copied from best practice guides, resource allocation that follows last year's budget, metrics that reward activity over outcomes.
The result? A system designed by accident. One that worked when markets were predictable, roles were stable, and change happened slowly. One that's now visibly breaking under the weight of weekly pivots, hybrid work, and the emergence of AI as a collaborative force.
So what actually is an operating model?
At its simplest: an operating model is how an organisation creates and delivers value. More precisely, it's the architecture of decisions, resources, governance, metrics, and ways of working that connect strategic intent to frontline execution.
It’s not your business model (what value you create for whom), your process model (how tasks get done), or your org chart (who reports to whom). It’s the connective tissue between them, the pattern of flows that determines whether your strategy ever becomes reality.
McKinsey's 2025 research identifies twelve interconnected elements in modern operating models, from purpose and value agenda through to technology and talent.[^1] But at the core, every operating model design comes down to five critical choices:
Who owns what - clarity on decision rights and accountability
When and how often - the rhythm of governance and adaptation
Where resources go - how funding and people flow to priorities
What gets tracked - the metrics that signal success or failure
How collaboration happens - the rituals, tools, and norms that enable work
Get these five elements aligned, and you create a system where strategy flows into execution with minimal friction. Get them misaligned, and you get the opposite: slow decisions, trapped resources, innovation theatre, and exhausted teams achieving mediocre results.
The challenge in 2025 is that most operating models were designed for a world that no longer exists. They assume stability, co-location, and human-only execution. The new reality demands models that can adapt at speed, orchestrate hybrid teams, and increasingly collaborate with autonomous AI agents.
Before you can evolve your operating model, you need to understand which archetype you're currently running, and whether it's capable of delivering the strategy you've set.
Why Most Operating Models No Longer Work
The statistics are sobering. McKinsey's 2025 research across 2,000 executives reveals that organisations typically lose 20-30% of their potential returns on capital due to poor operating model alignment, with the gap between strategic intent and delivered performance widening in volatile markets.[^1]
PwC's analysis of 245 executives in energy and resources sectors found similar patterns: even high-performing companies leave significant value on the table, not because their strategies are wrong, but because the invisible architecture connecting strategy to execution is broken.[^2]
The transformation landscape tells the same story. About 70% of business transformations fail to achieve their intended objectives.[^3] That's not a rounding error, it's a structural problem. When seven out of ten change initiatives miss their targets, the issue isn't execution discipline or change fatigue. It's that organisations are trying to transform without addressing the operating model that determines how transformation actually happens.
McKinsey identifies six frequent failure modes that plague operating model transformations:[^4]
Lack of behavioural readiness - teams aren't equipped for new ways of working
Poor incentive alignment - metrics and rewards still point to old behaviours
Insufficient depth in pilot conversion - surface changes that don't stick
Under-resourced mobilisation - transformation treated as a side project
Lack of adaptive capacity - inability to adjust as reality unfolds
Ignoring organisational "fit" - copying models without considering context
Each failure mode traces back to the same root cause: treating the operating model as structure (boxes and lines) rather than system (flows and rhythms).
The cost of getting this wrong goes beyond missed targets. Organisations report direct consulting and IT outlays of $10-25 million for large failed transformations, plus the opportunity cost of 20-30% lost ROCE (return on capital employed), declining employee engagement, and customer experience stagnation.[^5] More damaging than the financial cost is the organisational scar tissue, cynicism about change, erosion of trust in leadership, and the talent drain as your best people leave for organisations that work better.
Why Now? Why 2025 Is Different
Three forces are converging to expose the limits of legacy operating models:
Pace of change has outstripped governance rhythms
Customer preferences shift weekly. Technology capabilities evolve monthly. Competitors launch in weeks. Yet most organisations still run quarterly business reviews and annual planning cycles. By the time governance catches up, the decision is obsolete.Hybrid work has shattered the illusion of informal coordination
In the office, ambiguous decision rights and unclear processes were papered over by hallway conversations and ad hoc problem-solving. Hybrid work removed that safety net. Without intentional operating design, work simply doesn't flow.AI is redefining what "execution" means
AI isn't just automating tasks, it's becoming a collaborative capability that can sense, decide, and act with increasing autonomy. Operating models built around human-only execution are being disrupted by the emergence of human-AI teams. Organisations that can't redesign for this reality will find themselves competing with one hand tied behind their back.
The good news: success rates are improving. McKinsey's 2025 survey shows that 63% of operating model redesigns now meet most of their objectives, up dramatically from just 21% a decade ago. The difference? Leaders are finally treating operating model design as a holistic system rather than a structural shuffle.[^6]
But that still means more than one in three transformations fail. The question isn't whether your operating model needs redesigning, it's whether you'll do it intentionally before the market forces your hand.
The Operating Model Spectrum
Operating models exist on a spectrum, from hierarchies designed for control to networks designed for learning. No single model is "best", but understanding where yours sits and where it needs to evolve is essential.
Traditional Models: Built for Control, Struggling With Speed
Traditional operating models still dominate the corporate landscape. McKinsey's 2025 research shows that 89% of organisations continue to use one of three traditional structures: functional hierarchies, matrix management, or business unit models.[^7]
These models were built for a different era, when markets moved slowly, customer expectations were stable, and competitive advantage came from scale and efficiency. They excel at creating order, consistency, and control. But in 2025, their limits are showing.
Functional Model
Organises around specialist departments (Finance, HR, Marketing, Operations).
Strengths: Depth of expertise, process consistency, economies of scale
Limits: Slow cross-functional collaboration, limited end-to-end visibility
Best fit: Stable industries, government, large shared-service environments
Matrix Model
Combines functional and project/regional reporting to share resources across priorities.
Strengths: Balanced perspective across geographies and products
Limits: Dual accountability, decision ambiguity, time-consuming governance
Best fit: Complex, multinational enterprises where cross-coordination is essential
Business Unit Model
Divides the organisation into semi-autonomous divisions, each with its own P&L.
Strengths: Speed and responsiveness to local markets
Limits: Duplicated effort, fragmented culture, weak enterprise coherence
Best fit: Diversified portfolios or conglomerates serving different customer sets
Notice how control gives way to coherence as you move right along the spectrum, from functional silos to adaptive networks.
The tension between control and speed defines the transition to modern models.
Emerging Models: Built for Flow and Responsiveness
As volatility and digital interdependence increased, organisations began re-architecting around value flow instead of functional hierarchy. These emerging designs, representing 11% of organisations but growing rapidly, seek responsiveness, coherence, and customer centricity.[^8]
Product-Platform Model
Teams organised around products or platforms, with cross-functional groups owning outcomes end-to-end.
Traits: Customer-focused metrics, continuous delivery, clear ownership
Best fit: Digital businesses, modern service organisations seeking persistent accountability
Leadership shift: From project funding to product investment; from milestone tracking to value tracking
Enterprise-Agile Model
Extends agile principles beyond technology teams into enterprise governance.
Traits: Multidisciplinary squads/tribes aligned to value streams; iterative planning; faster governance loops
Best fit: Sectors needing rapid innovation, banking, telecommunications, digital government
Leadership shift: From command-and-control to empowerment and transparency
Example: A global financial services firm replaced its matrix with an enterprise-agile model and reported a five-to-tenfold increase in decision speed alongside double-digit gains in customer satisfaction and operational performance.[^9] The shift wasn't just structural, it redesigned decision rights, governance cadence, and ways of working as a coherent system.
Ecosystem / Network Model
Recognises that value creation increasingly crosses organisational boundaries.
Traits: Partnerships, shared data platforms, co-innovation orchestrated through mutual trust
Best fit: Industries where collaboration drives advantage, fintech, sustainability, trade
Leadership shift: From ownership to orchestration
Across emerging models, decision-making moves closer to customers, governance rhythms accelerate, and collaboration replaces hierarchy as the organising logic.
AI-Native Models: Built for Learning and Autonomy
Artificial intelligence introduces a third era of operating model design. If the 20th century was about control and the early 21st about adaptability, the next phase is about organisational intelligence, systems that sense, decide, learn, and act continuously.
What Makes an Organisation "AI-Native"?
An AI-native organisation isn't one that uses AI tools. It's one where AI capabilities are embedded in the core operating architecture, fundamentally reshaping how value is created and decisions are made. McKinsey describes this shift from traditional "org charts" to "work charts", diagrams that map the exchange of tasks and outcomes between humans and AI agents, rather than hierarchical reporting lines.[^10]
By 2025, 78% of organisations report using AI technologies, up from 55% the previous year.[^11] But adoption doesn't equal transformation. True AI-native organisations are rarer, and their performance advantage is stark.
McKinsey and MIT's 2025 study shows that AI-native leaders achieve 3.8x KPI improvements over laggards, with payback periods shortening to 6-12 months.[^12] The gap isn't closing, it's widening. Organisations that fully industrialise agentic architectures are securing sizable business gains, while late adopters accumulate technical debt and face regulatory hurdles.
The Agentic Organisation
The most advanced AI-native model is what McKinsey calls the "agentic organisation", where autonomous AI agents perform routine analysis and decisions within defined guardrails, allowing humans to focus on creativity, ethics, and sense-making.[^13]
Key characteristics:
Autonomous agents handle workflows end-to-end (e.g., revenue operations, IT incident resolution) often without human intervention
Real-time governance replaces periodic reviews, AI proposes budgets, runs forecasts, provides continuous insights
Embedded algorithms self-optimise within workflows, learning from outcomes
Human oversight shifts from direct control to strategic guidance and system design
Real-World Evidence
The shift is already happening:
Careem (Dubai ride-hailing) flattened corporate hierarchy through deep AI integration across operational management and customer service[^14]
Lloyds Banking Group uses Vertex AI to streamline ML experimentation across 300+ data scientists, reducing income verification in mortgage applications from days to seconds[^15]
Toyota implemented an AI platform enabling factory workers to develop and deploy ML models, reducing over 10,000 man-hours annually[^16]
A McKinsey–MIT study (2025) found that AI-native leaders achieved 3.8× KPI improvements and recouped investment within 6–12 months [^17]
The Governance Challenge
AI-native models demand new governance structures. Gartner ranks agentic AI as the #1 strategic trend for 2025 but predicts 40% of early enterprise projects could be cancelled by 2027 due to governance gaps.[^18] The challenge isn't technical, it's organisational. How do you govern agents that operate 24/7? How do you ensure accountability when AI makes autonomous decisions? How do you maintain ethical standards at machine speed?
Many organisations have established Chief AI Officer (CAIO) roles, AI advisory boards, and agentic governance councils responsible for oversight, risk management, and strategic alignment.[^19] Best practices include:
Agentic governance councils with clear oversight and risk management protocols
Enterprise-level KPIs tracking autonomy ratio and decision speed
Real-time monitoring replacing traditional audit cycles
Ethical guardrails encoded at the system level, not bolted on afterward
Beyond Structure: The Ambidextrous Operating Model
Perhaps the most sophisticated insight comes from emerging practice: organisations are no longer standardising on a single operating model. Instead, they’re running multiple models simultaneously.
Research by O’Reilly and Tushman [^20] describes this as organisational ambidexterity: the ability to explore new opportunities while exploiting existing strengths at the same time.
For example:
R&D might run a "Studio" model optimised for experimentation
Core products or operations might use a "Tribe" model focused on enterprise agility
Internal platforms might operate as shared services with clear SLAs
AI-enabled customer service might run an agentic model with minimal human oversight
This approach recognises that different parts of the organisation move at different speeds. The goal isn’t structural uniformity, it’s maintaining strategic coherence across diverse models. The coordination challenge is real.
Gartner’s [^21] research on Digital Twins of an Organisation (DTOs) points to an emerging solution: dynamic software models that let leaders see how the organisation actually operates, simulate changes, and allocate resources more effectively across varied contexts.
While still maturing, DTO platforms, combined with AI-driven analytics, offer a way to manage complexity that would overwhelm traditional governance structures.
The Provocative Truth
The AI-native organisation isn't a re-wired version of the old model. It's a fundamentally different paradigm, one where the operating model itself becomes adaptive, learning, and increasingly autonomous in making routine decisions, while humans focus on orchestration, learning, and ethical oversight.
But most organisations aren't ready for this. They need to choose their next step, not their final state.
Choosing and Designing the Model That Fits
There is no single best operating model, only one that best fits your strategy, pace, and risk appetite. The key is coherence: structure, decision-making, governance, and ways of working that move value at the speed your market demands.
Choosing your operating model isn't about picking the most modern option. It's about honest assessment of where you are, where you need to go, and what change capacity you actually have.
Three Questions to Guide Your Choice
1. Strategy: What speed of adaptation does your strategy demand?
If your strategy relies on stability and efficiency (regulated industries, mature products, asset-heavy operations), traditional functional or business unit models may still serve you well.
If your strategy demands rapid iteration and customer responsiveness (digital services, consumer tech, banking innovation), emerging models like product-platform or enterprise agile are the better fit.
If your strategy requires continuous learning and autonomous execution at scale (data-rich operations, AI-enabled services, real-time personalisation), you need to build toward AI-native models.
The honest question: Is your current operating model capable of executing your strategy at the speed your competitors are moving?
2. Maturity: Where's your digital and data maturity actually at?
Low maturity (siloed data, manual processes, limited digital capabilities): Focus first on building foundational capabilities. Don't leap to enterprise agile or AI-native models without the technical substrate to support them.
Medium maturity (some digital platforms, emerging data capabilities, pockets of agile practice): You're ready to pilot emerging models in select areas while modernising your core.
High maturity (integrated data platforms, strong digital capabilities, agile at scale): You can credibly design for AI-native operating models and experiment with agentic approaches.
The honest question: Are you trying to run an AI-native playbook on legacy infrastructure?
3. Culture: What change capacity does your culture have?
Low appetite for change (risk-averse, hierarchical, change fatigue): Incremental evolution within your current model may be the pragmatic path. Fix the five levers; decision rights, governance cadence, resource allocation, metrics, and ways of working, before attempting wholesale structural change.
Medium appetite (some agility, growing innovation culture, selective risk-taking): You can pilot emerging models in specific business units or value streams while maintaining traditional structures in stable areas.
High appetite (innovation-driven, comfortable with ambiguity, learning orientation): You're positioned to lead with AI-native experimentation and fractalized operating models.
The honest question: Are you designing for the culture you want or the culture you have?
Designing for Hybrid Reality
Most organisations blend elements from multiple models, and that’s not failure, it’s design pragmatism. The goal isn’t purity; it’s coherence, knowing which model fits where, and why.
A global bank might run:
Traditional functional shared services for compliance and finance
Enterprise agile squads for digital customer experience
Product-platform models for core banking technology
Emerging agentic pilots for customer service automation
The goal isn't purity. It's intentional design. Understand which model you're running where, why it fits that context, and how the pieces connect to create (not destroy) value across the enterprise
To choose your operating model:
Name your current model honestly. Functional? Matrix? Product? Hybrid? Don't describe the model you announced, describe the one that's actually running.
Identify your pain points. Where is value getting stuck; decisions, resources, information flow, or culture?
Match design to environment. Stable markets reward efficiency. Dynamic markets reward adaptability. AI-enabled ecosystems reward intelligent, autonomous execution.
Start where you have permission. If enterprise-wide change feels impossible, pick one value stream or business unit and redesign it as a proof point.
Iterate intentionally. Treat operating model design as a living discipline, not a one-off project. McKinsey's data shows the most successful organisations revisit and refine their models every 2–3 years.[^22]
The Next Step: How to Design and Tune
Once you've chosen your target operating model, the real work begins: designing the five levers; decision rights, governance cadence, resource allocation, metrics, and ways of working, to make that model real.
Map It: The 90-Minute Operating-Model Diagnostic
You can't redesign what you can't see. Most leadership teams have never explicitly mapped their operating model, they've just inherited and modified it over time. This 90-minute diagnostic surfaces what's actually running beneath your org chart.
The Operating Model Clarity Session
Block 90 minutes with your senior team. No laptops, no slide decks. Just a whiteboard and willingness to examine how work truly flows in your organisation.
Work through these five diagnostic questions:
Question 1: Where do critical decisions actually get made?
Pick your three most important recent decisions (product direction, major hiring, technology platform choice). For each:
Who made the call? (Not who had input, who decided.)
How long from question to decision?
How many meetings, emails, or escalations?
Did the right information reach the decision-maker?
If you can't name a single owner for critical decisions, or if the path from question to answer involves more than three steps, you have a decision rights problem.
Question 2: What's your governance rhythm, and does it match your market?
Map your key governance forums: board meetings, exec committees, portfolio reviews, planning cycles. For each, note:
How often does it meet?
What authority does it actually have? (Information-sharing or decision-making?)
How much lead time to get something on the agenda?
What happens between meetings when priorities shift?
If your most frequent decision forum meets less often than your market changes, you're governing in the rearview mirror.
Question 3: Can you follow the money?
Take your top three strategic priorities. Now trace:
What percentage of your budget is allocated to each?
When was that allocation decided?
How quickly could you shift resources if one priority became urgent?
What would it take to stop funding something that's not working?
If strategic priorities and budget allocation don't match, or if reallocation takes longer than one quarter, your resource system isn't serving your strategy.
Question 4: What signals success in your organisation?
Ask your exec team (separately, not in the room) to name the top three metrics that determine whether the business is winning. If you get inconsistent answers, you have an alignment problem.
Then ask: Do those metrics measure what customers care about, or what's easy to count? Activity metrics (features shipped, hours logged, utilisation rates) reveal a system optimising for busyness. Outcome metrics (customer lifetime value, problem resolution, adoption rates) reveal a system optimising for value.
Question 5: How does information actually travel?
Pick a customer insight that recently changed something important (a complaint, a feature request, a churn signal). Trace its path:
Where was it captured?
Who saw it first?
How did it reach someone with authority to act?
How long did that journey take?
If valuable information takes more than one week to reach decision-makers, your coordination system is creating drag.
What the Patterns Tell You
After 90 minutes, look for recurring themes:
Everything escalates up → Decision rights need pushing down
Meetings don't conclude → Governance forums lack authority or information
Budget is unmovable → Resource allocation is annual, not adaptive
Success is ambiguous → Metrics aren't aligned to outcomes
Information gets stuck → Ways of working create silos or handoff friction
The First Action: Pick One Friction Point
Don't attempt enterprise-wide transformation. Pick one system where value is visibly stuck. Maybe it's:
Decision bottlenecks in product development
Quarterly governance that can't respond to monthly changes
Budget rigidity that traps resources in yesterday's priorities
Metrics that reward the wrong behaviours
Coordination friction across teams serving the same customer
Fix that one point. Prove the system can change. Build leadership alignment around a different way of working. Then cascade to the next friction point.
What Success Looks Like
McKinsey's 2025 research is encouraging: 63% of operating model redesigns now achieve most of their objectives, triple the success rate from a decade ago.[^23] The shift? Leaders stopped treating this as a one-time restructure and started treating it as intentional system design.
Successful organisations don't copy models from exemplars. They:
Understand which archetype fits their strategy and context
Design the five core elements (decisions, governance, resources, metrics, ways of working) as a coherent system
Iterate based on what they learn, not defend what they designed
Revisit and refine every 2-3 years as strategy and environment evolve
Your operating model is running exactly as designed - even if it's designed to be slow, siloed, and reactive. The question is whether you'll make the invisible visible, then redesign it intentionally.
In 2025, competitive advantage no longer comes from better strategy. It comes from better systems that turn strategy into execution with speed, clarity, and continuous adaptation.
The organisations that win will be those that can see their operating model clearly, and redesign it before the market forces their hand.
Researched and written by Rebecca Agent, with credit to the following AI tools for assistance in producing this content:
Editorial and grammar writing assistant | Grammarly (English US)
Research, writing, reader timing and SEO | ChatGPT, and Claude
The Deep Dive Podcast Overview | NotebookLM by Google
Topic research to link peer-reviewed research papers | Storm Genini Stanford; Google Gemini, Perplexity
. . .
Core Research
McKinsey & Company (2025). A New Operating Model for a New World.
McKinsey & Company (2025). The Agentic Organization: Contours of the Next Paradigm for the AI Era.
Strategy& (PwC). The Strategic Operating Model.
REFERENCES
[^1]: McKinsey & Company. (2025). "A new operating model for a new world.“
[^2]: PwC. "Performance Alignment."
[^3]: ProcureInsights. (2025). "McKinsey's New Framework Is Not the Answer."
[^4]: McKinsey & Company. (2025). "How to get your operating model transformation back on track."
[^5]: ProcureInsights. (2025). "McKinsey's New Framework Is Not the Answer."
[^6]: McKinsey & Company. (2025). "The new rules for getting your operating model redesign right."
[^7]: McKinsey & Company. (2025). "A new operating model for a new world.“
[^8]: McKinsey & Company. (2025). "A new operating model for a new world.”
[^9]: McKinsey & Company. (2025). ""A new operating model for a new world.“" Case study: Global financial services infrastructure company.
[^10]: McKinsey & Company. (2025). "The agentic organization: contours of the next paradigm for the AI era."
[^11]: Timspark. (2025). "The AI Evolution: Past, Present & Future."
[^12]: Enterprise AI Executive. (2025). "McKinsey-MIT Reveal Major AI Performance Gap."
[^13]: McKinsey & Company. (2025). "The agentic organization."
[^14]: Oracle. (2025) "Careem increases efficiency and cuts invoice process time 70% with Oracle AI."
[^15]: Google Cloud. (2025) “Lloyds Banking Group: Improving customer experience with machine learning”
[^16]: Google Cloud. (2024) “Toyota shifts into overdrive: Developing an AI platform for enhanced manufacturing efficiency”
[^17]: Enterprise AI Executive "McKinsey-MIT reveal major AI performance gap”
[^18]: Gartner. (2025) “Gartner Predicts Over 40% of Agentic AI Projects Will Be Canceled by End of 2027“
[^19]: McKinsey & Company. (2025). "The State of AI."
[^20]: O’Reilly, C. A., & Tushman, M. L. (2013). “Organizational Ambidexterity: Past, Present, and Future.” Academy of Management Perspectives, 27(4), 324–338.
[^21] Gartner Research. (2024). “Digital Twin of an Organization: Use Cases and Maturity Report.”
[^22]: McKinsey & Company. (2025). "A new operating model for a new world."
[^23]: McKinsey & Company. (2025). "The new rules for getting your operating model redesign right."
Core Concepts:
Operating Model:
The system that connects strategy to execution — how decisions, resources, and information flow through an organisation to create value.McKinsey’s Twelve Elements:
A 2025 framework showing how purpose, governance, talent, technology, and value delivery integrate to create a cohesive system — proving that operating models are built, not inherited.Decision Rights:
Clear ownership of who decides what, and how fast — the foundation of speed, accountability, and coherence.Governance Cadence:
The rhythm of sensing, deciding, and adjusting — aligning decision cycles with the pace of market change.Resource Allocation:
How funding and people flow to priorities — the mechanism that signals what truly matters.Metrics That Drive Behaviour:
Outcome-based measures that focus on value creation, not activity — enabling self-correction and alignment.Ways of Working:
The rituals, norms, and tools that shape how collaboration and learning actually happen day to day.Five Levers of Design:
Decision rights, governance cadence, resource allocation, metrics, and ways of working — the practical system levers that make strategy executable.Transformation Failure Modes:
Common pitfalls identified in McKinsey’s research — from behavioural unreadiness and misaligned incentives to under-resourced mobilisation and lack of adaptive capacity.Operating Model Archetypes:
Distinct design patterns along a spectrum from traditional (functional, matrix, business unit) to emerging (product-platform, agile, ecosystem) to AI-native models.Organisational Ambidexterity:
The ability to run multiple models at once — exploring new opportunities while exploiting existing strengths.Digital Twin of the Organisation (DTO):
A dynamic digital model that mirrors real operations — helping leaders test scenarios, monitor performance, and optimise resource flow.AI-Native Organisation:
A model where AI is embedded in the operating architecture — enabling systems that sense, decide, and act autonomously, while humans focus on orchestration and ethics.Agentic Organisation:
An advanced AI-native design where autonomous agents execute defined workflows and decisions within ethical guardrails, supervised by humans.Systemic Design:
Treating the operating model as a living system — tuned for flow, learning, and adaptability, rather than static structure or hierarchy.
. . .
See related pieces (on the blog) :
If this content was helpful to you and your work please comment, share, and like !