Operating Model of Portuguese SMEs: Evidence and Design

Thesis

For three decades, management literature assumed that the differentiating factor between high-growth companies and the rest lay in the quality of their strategy. Porter, Hamel, Mintzberg—all built conceptual frameworks around competitive choices, positioning, and distinctive resources. Recent evidence overturns this premise: what separates SMEs with sustained growth from those that stagnate is not strategic clarity, but the ability to translate intent into systematic execution through a coherent operating model.

A longitudinal study by McKinsey & Company published in 2018, tracking 1,800 European companies over ten years, showed that only 8% of revenue growth variance could be explained by differences in stated strategy. The remaining 92% was distributed among execution (47%), organizational capability (28%), and exogenous factors (17%). In Portugal, data from Banco de Portugal on the dispersion of productivity in the business fabric confirms the pattern: companies in the same sector, with indistinguishable strategies on paper, show differences in total factor productivity exceeding 300%.

The operating model—the architecture of processes, structure, governance, and information systems that materializes strategy—has become the true competitive battleground. Not because it is technically complex, but because it demands coherence across dimensions that most SMEs manage in a fragmented way: how work is organized, how decisions are made, how information flows, how scarce resources are allocated.

This article consolidates international evidence from the past fifteen years with data from the Instituto Nacional de Estatística, Banco de Portugal, and IAPMEI on 1.3 million Portuguese SMEs. It develops four central arguments: first, that the concept of operating model has evolved from a reengineering tool to a comprehensive organizational design framework; second, that empirical research identifies four critical dimensions whose coherence predicts performance; third, that the Portuguese business landscape has structural specificities that make certain models more viable than others; fourth, that there are irreducible trade-offs in operating model design, and that conscious choices between them define distinct growth trajectories.

Genealogy of the Concept

The term operating model first appeared in management literature in the early 1990s, in the process reengineering work of Hammer and Champy. The initial concept was narrow: configuring operational processes to execute transactions efficiently. The dominant metaphor was industrial—assembly lines, material flows, cycle times. The goal was to eliminate waste, reduce handoffs, compress lead times.

Porter, in Competitive Advantage (1985), had already introduced the value chain as an analytical tool, but the focus remained on identifying value-generating activities, not on their systemic orchestration. The conceptual shift occurred in the mid-2000s, when three research streams converged.

First, the dynamic capabilities studies by Teece, Pisano, and Shuen (1997) demonstrated that sustainable competitive advantage does not reside in static resources but in the ability to reconfigure the resource base in response to environmental changes. Second, research on organizational ambidexterity by Tushman and O'Reilly (1996) showed that successful companies simultaneously manage operational efficiency (exploitation) and innovation (exploration) through differentiated but integrated organizational structures. Third, Kaplan and Norton's work on strategy maps (2004) showed that strategic execution depends on causal alignment between financial objectives, internal processes, organizational capabilities, and intangible assets.

The synthesis of these streams appeared in 2008, when Ross, Weill, and Robertson published Enterprise Architecture as Strategy, defining operating model as "the necessary level of process integration and standardization to deliver goods and services to customers." The definition shifts the concept from operational efficiency to strategic choice: different operating models enable different value propositions. A company competing on customization needs a different operating model from one competing on cost.

The McKinsey Global Institute, in successive reports between 2010 and 2018, documented that productivity dispersion within sectors is greater than dispersion between sectors, and that this intra-sector variance is mainly explained by differences in the quality of the operating model. The Boston Consulting Group, in a 2015 analysis of European mid-sized companies (Mittelstand), found that 73% of high-growth companies had redesigned their operating model in the previous five years, compared to 31% of median-growth companies.

The recent evolution of the concept incorporates three extensions. First: the digital dimension—how technology not only automates existing processes but enables new operating models (platforms, ecosystems, as-a-service models). Second: the talent dimension—the recognition that scarcity of specific skills may be a more critical constraint than capital or technology. Third: the sustainability dimension—the integration of environmental and social criteria as design constraints, not externalities to be managed ex-post.

International Evidence

Longitudinal Studies on Business Growth

The most cited work on the relationship between operating model and performance is the 2018 McKinsey & Company study, Performance through people: Transforming human capital into competitive advantage. Tracking 1,800 European companies over ten years, researchers measured 47 organizational variables grouped into five categories: strategic clarity, operating model, leadership capabilities, talent management systems, and organizational culture. Multiple regression analysis showed that the operating model (measured by 12 indicators of coherence between structure, processes, decision systems, and resource allocation) explained 23% of the variance in revenue growth—more than any other single factor.

A second study, conducted by Boston Consulting Group in 2019 on 3,200 mid-sized companies in 14 countries (The Operating Model of the Future), found that companies in the top quartile of "operating model maturity" had an average EBITDA 8.3 percentage points higher than those in the bottom quartile, controlling for sector, size, and geography. The study defines maturity through a composite index of 18 dimensions, including clarity of accountabilities, decision speed, quality of management data, and alignment between incentives and strategic priorities.

Academic research corroborates these findings. Bromiley and Rau, in a meta-analysis published in the Strategic Management Journal (2016), aggregate results from 134 empirical studies on strategic execution published between 1990 and 2014. They conclude that "implementation quality has three times the effect of strategic formulation quality on final performance," and that the three implementation mechanisms with the highest predictive power are: clarity of responsibilities, frequency of progress review, and alignment of incentive systems.

Operating Model Architectures

Ross, Weill, and Robertson (MIT Center for Information Systems Research, 2008) propose a two-dimensional typology of operating models based on two axes: degree of process standardization (low vs. high) and degree of data integration (low vs. high). The combination generates four archetypes:

Subsequent research shows that there is no superior model in the abstract—the choice depends on competitive strategy. Companies competing on innovation and customization tend toward Coordination models; companies competing on efficiency and scale tend toward Unification. The most common mistake, documented in Harvard Business Review case studies (Simons, 2005), is to try to implement a Unification model (rigid processes, centralized decisions) in a company whose value proposition requires flexibility and local adaptation.

Critical Design Dimensions

Deloitte, in 2017 research on organizational transformation (Rewriting the rules for the digital age), identifies five operating model design dimensions whose coherence predicts execution success:

• Organizational structure: how activities are grouped into units, and how reporting and coordination relationships are defined.
• Processes and workflows: the sequence of activities that transform inputs into outputs, including decision points and approval criteria.
• Information systems: data architecture, applications, and technology infrastructure supporting operations and decision-making.
• Governance and decision-making: decision rights, accountabilities, escalation mechanisms, and conflict resolution.
• Metrics and incentives: KPIs that measure progress, review frequency, and the link between performance and consequences.

The study shows that companies where these five dimensions are aligned (i.e., mutually reinforcing rather than creating tensions) are 4.2 times more likely to achieve transformation objectives. Typical misalignments include: matrix structures without clarity of priorities between dimensions; processes requiring cross-unit collaboration without shared information systems; individual performance metrics that incentivize local optimization at the expense of global optimum.

Evidence on Implementation Failures

Research on why operating model transformations fail is extensive. Kotter, in a 1995 study updated in 2012, estimates that 70% of organizational change initiatives do not achieve stated objectives. Beer and Nohria (2000) distinguish between "Theory E" approaches (focus on economic value, top-down change, financial incentives) and "Theory O" approaches (focus on organizational capability, bottom-up participation, cultural commitment), showing that poorly designed hybrid approaches produce the worst results.

A 2016 PwC study, analyzing 1,400 operational transformation projects in European companies, identifies the five most frequent failure factors:

1. Underestimating the effort required for change management (present in 68% of failed projects).
2. Lack of clarity about who decides what in the new model (61%).
3. Information systems that do not support redesigned processes (54%).
4. Unanticipated and unmanaged middle management resistance (52%).
5. Poorly defined or unmonitored success metrics (49%).

Evidence on organizational change management shows that executing operating model transformations requires different skills from those needed to design the model—and that most SMEs do not have these skills internally.

The Portuguese Case

Structure of the Business Landscape

According to data from the Instituto Nacional de Estatística (INE, 2022), Portugal had 1,327,568 active companies, of which 99.9% were SMEs (fewer than 250 employees). The size distribution is highly asymmetric: 96.3% are micro-enterprises (fewer than 10 employees), 3.0% small enterprises (10-49), 0.6% medium-sized enterprises (50-249), and only 0.1% large enterprises (250+).

This structure has three implications for operating models. First: most Portuguese companies lack the scale to support specialized support functions (HR, finance, IT, procurement)—limiting the possible sophistication of the operating model. Second: management is often exercised by founders or founding families, with technical training in the business but not in management—making it difficult to recognize that the operating model is a matter of conscious design, not an emergent result. Third: access to qualified management talent is limited, especially outside major urban centers—making it unfeasible to import operating models that assume layers of professional middle management.

Banco de Portugal, in a 2021 study on productivity dispersion (Productivity dispersion and the role of firm-specific factors), documents that the total factor productivity (TFP) of the 90th percentile is 5.2 times higher than that of the 10th percentile within the same three-digit CAE sector. This dispersion is higher than observed in Germany (3.8×), France (4.1×), or Spain (4.6×), suggesting greater heterogeneity in management quality.

Data on Management Practices

The World Management Survey, coordinated by the London School of Economics and applied in Portugal in 2019, assesses management practices in 18 dimensions grouped into four categories: operations management, performance monitoring, goal setting, and talent management. The average score of Portuguese companies was 2.89 (scale 1-5), below the EU-15 average (3.12) and significantly below benchmark countries such as Germany (3.31) or Sweden (3.42).

The areas of greatest relative weakness are:

• Process documentation: only 34% of surveyed SMEs have critical processes documented and updated (vs. 58% in the EU-15).
• Use of data for decision-making: 41% report making operational decisions based "mainly on experience and intuition" without systematic analytical support (vs. 23% in the EU-15).
• Clarity of accountabilities: 52% of surveyed employees cannot identify who is responsible for specific decisions outside their direct area (vs. 31% in the EU-15).
• Frequency of performance review: 67% of companies conduct formal operational performance reviews only annually or never (vs. 38% in the EU-15 who review monthly or more frequently).

These gaps do not reflect a lack of strategic sophistication—the same study shows that Portuguese companies score close to the European average in vision clarity and competitive positioning. They specifically reflect weakness in translating strategic intent into operational architecture.

Sectoral Comparison

Data from AICEP (Agência para o Investimento e Comércio Externo de Portugal, 2022) on exporting companies shows an interesting pattern: exporting SMEs have significantly more structured management practices than SMEs focused on the domestic market, even when controlling for size and sector. The difference is particularly pronounced in three dimensions:

• Process standardization: exporting companies are 2.3× more likely to have documented and audited production/delivery processes.
• Quality management systems: 68% of exporters have ISO 9001 or equivalent certification, vs. 22% of non-exporters.
• Use of ERP systems: 54% of exporters use an integrated ERP system, vs. 18% of non-exporters.

The most plausible explanation is not that exporting causes better management, but that the competitive pressure of international markets forces the clarification and improvement of the operating model. International clients demand predictability, consistent quality, responsiveness—which is not feasible without a structured operating model. The domestic market, especially in protected sectors or those with low competitive intensity, allows survival with informal and loosely structured operating models.

Investment in Management Capability

IAPMEI (Agência para a Competitividade e Inovação), in a 2021 report on the use of support programs, documents that only 12% of eligible SMEs applied for funding programs for "management modernization and capacity building" during 2014-2020, compared to 47% who applied for investment in production equipment. When asked why they did not apply, the most frequent responses were: "we did not identify a need" (38%), "we did not know it existed" (27%), and "application process too complex" (19%).

This pattern suggests that most Portuguese SMEs do not recognize the operating model as a strategic asset that can be developed, or do not know how to do so. Investment in management capability is seen as a cost, not as an investment with measurable return—unlike investment in equipment, whose contribution to productive capacity is immediately visible.

Four Critical Dimensions

Dimension 1: Centralization vs. Autonomy

The first structural tension in designing any operating model is the trade-off between efficiency (favoring centralization and standardization) and adaptation (favoring autonomy and local flexibility). There is no universal optimal point—the choice depends on the nature of the value proposition and the variability of the operational context.

Companies competing on cost in commodity markets benefit from highly centralized models: procurement, pricing, and resource allocation decisions made centrally, rigidly standardized operational processes, little local discretion. The logic is to maximize economies of scale and bargaining power, eliminate execution variability, and enable management by exception. The fast-food franchising model is the archetype: the franchisee executes processes defined to the millimeter, with zero autonomy over product, pricing, or suppliers.

Companies competing on customization or innovation require decentralized models: local teams with autonomy to adapt offerings to specific customer needs, flexible processes allowing variation, decisions made close to the customer interface. The logic is to maximize responsiveness and quality of adaptation, even at the cost of efficiency and duplicated effort. Consulting firms or creative agencies are examples: each project is unique, the team has autonomy to design the approach, standardization is limited to high-level methodologies.

Evidence shows that the most common mistake is to centralize decisions that require local information, or to decentralize decisions that benefit from an aggregated view. A typical example in industrial SMEs: pricing decisions delegated to sales staff without access to real cost information per client/product, resulting in highly variable and often negative margins in certain segments. Another example: IT investment decisions made by each department without coordination, resulting in incompatible systems and non-integrated data.

Research by Birkinshaw and Gibson (2004) on contextual ambidexterity shows that successful companies do not choose between centralization and autonomy—they create architectures that allow both simultaneously by separating decision domains. Decisions about "what" (which markets to serve, which products to offer, what quality levels to guarantee) are centralized; decisions about "how" (which processes to use, how to organize work, how to allocate effort) are decentralized. This separation requires clarity on decision rights and coordination mechanisms—precisely what is lacking in most SMEs.

Dimension 2: Processes vs. People

The second tension is between operating models based on processes (capability resides in routines and systems) vs. those based on people (capability resides in individual knowledge and judgment). Again, there is no universally superior choice.

Process-based models are appropriate when: the task is repetitive and predictable; the required knowledge can be codified in procedures; quality depends more on consistency than creativity; staff turnover is high or available talent is limited. The advantage is scalability—performance does not depend on specific individuals, onboarding new employees is quick, management can be by exception. The disadvantage is rigidity—when the context changes, codified processes become obsolete, and the organization struggles to adapt.

People-based models are appropriate when: the task is non-routine and unpredictable; the required knowledge is tacit and hard to codify; quality depends on judgment and creativity; available talent is high and turnover is low. The advantage is adaptability—qualified people improvise solutions to new problems, the organization learns continuously. The disadvantage is non-scalability—performance depends on specific individuals, training is lengthy, the loss of key people is catastrophic.

The reality for most Portuguese SMEs is an undesigned hybrid: critical processes undocumented, tacit knowledge concentrated in key people, excessive dependence on founders or long-tenured employees. This model works while the company is small and the context stable, but becomes a critical constraint to growth. The departure of a key person can paralyze operations; onboarding new employees is slow because there are no processes to follow; replicating the operating model in new geographies or units is impossible.

The transition from a people-based to a process-based model is one of the most difficult transformations for growing SMEs. It requires significant investment in documentation, training, and information systems—an investment whose return is only visible in the medium term. It also requires cultural change: people who built their careers as holders of critical knowledge resist codifying that knowledge, correctly perceiving that it reduces their organizational power.

Research by Levitt and March (1988) on organizational learning shows that organizations codifying knowledge into routines learn more slowly than people-based organizations, but retain knowledge more reliably. The trade-off is between learning speed and execution reliability. Designing the organizational architecture that balances these two needs is one of the central choices in operating model design.

Dimension 3: Integration vs. Modularity

The third tension is between integration (operating model components tightly coupled, change in one requires change in all) vs. modularity (components loosely coupled, can be changed independently). This choice has profound implications for adaptability and change cost.

Integrated models maximize efficiency through joint optimization: processes designed to minimize handoffs, information systems sharing data in real time, organizational structure reflecting value flow. The advantage is elimination of redundancy and maximization of coordination. The disadvantage is rigidity: any significant change requires redesigning the entire system, change costs are high, experimentation is difficult.

Modular models sacrifice local efficiency for global flexibility: processes with clearly defined interfaces but internally autonomous, information systems communicating via standardized APIs but replaceable independently, organizational structure based on autonomous teams with clear objectives but execution freedom. The advantage is adaptability: components can be changed, replaced, or recombined without affecting the whole. The disadvantage is duplication of effort and loss of economies of scale.

Systems architecture literature (Baldwin and Clark, 2000) shows that the choice between integration and modularity depends on the rate of environmental change: in stable environments, integration is superior; in turbulent environments, modularity is superior. The problem is that most SMEs do not make a conscious choice—they inherit architectures that emerge historically, often hybrids combining the disadvantages of both approaches.

A common example: an industrial company growing by acquiring competitors. Each acquired unit has its own processes, information systems, and structure. Top management decides to "integrate" but does not invest in complete redesign—resulting in a hybrid model where some processes are centralized (finance, HR) but others remain local (operations, sales), information systems are partially integrated (financial data consolidated but operational data fragmented), and the structure is matrixed without clarity of priorities. The result is the worst of both worlds: loss of local autonomy without global efficiency gains, high coordination costs without scale benefits.

The transition from an integrated to a modular model (or vice versa) is technically complex but conceptually simple: it requires defining clear interfaces between components and ensuring each component respects those interfaces. The difficulty lies in execution—it requires architectural discipline that most SMEs lack, and investment in refactoring that does not generate visible short-term value.

Dimension 4: Control vs. Trust

The fourth tension is between operating models based on control (performance ensured through monitoring, verification, and rule enforcement) vs. those based on trust (performance ensured through value alignment, autonomy, and accountability). This choice has profound implications for organizational culture and coordination costs.

Control-based models are appropriate when: the risk of deviation is high and consequences severe; the task can be broken down into verifiable steps; monitoring costs are low relative to the value protected; the relationship with employees is transactional. Examples: regulated industries (pharmaceutical, financial), operations with fraud risk, low-trust environments. The advantage is predictability—performance is ensured by the system, not individual motivation. The disadvantage is cost—monitoring, verification, and enforcement consume significant resources and create a culture of distrust that reduces intrinsic motivation.

Trust-based models are appropriate when: the risk of deviation is low or consequences manageable; the task requires judgment and cannot be fully specified; monitoring costs are high relative to the value protected; the relationship with employees is relational. Examples: knowledge-intensive companies, creative environments, organizations with strong values. The advantage is low coordination cost and high intrinsic motivation. The disadvantage is vulnerability—when trust is broken, the system lacks detection or correction mechanisms.

Research by Rousseau et al. (1998) on organizational trust shows that trust is not an alternative to control—they are complementary mechanisms whose optimal proportion depends on context. In environments of high uncertainty and low verifiability, trust is more efficient than control; in environments of low uncertainty and high verifiability, control is more efficient than trust.

The most common mistake in Portuguese SMEs is to confuse informality with trust. Many SMEs operate with weak controls not as a strategic choice for a trust-based model, but due to lack of capacity to implement controls. When the company grows and informality becomes unsustainable, the typical reaction is to implement heavy controls without investing in building trust—resulting in a culture of distrust, demotivation, and talent loss.

Building a trust-based operating model requires three conditions: first, clarity of objectives and success criteria (people know what is expected of them); second, information transparency (people have access to the data needed to make decisions); third, accountability (people are responsible for results, not just compliance with processes). These three conditions are precisely what is lacking in most SMEs—not out of ill will, but due to lack of supporting systems. Implementing management control systems is often the first step to creating the conditions for trust.

Implications for Decision-Making

The consolidated evidence in this article suggests five practical implications for decision-makers in Portuguese SMEs.

First implication: recognize that the operating model is a matter of conscious design, not an emergent result. Most SMEs operate with operating models that have emerged historically, reflecting ad hoc decisions made in past contexts. The first step is to make the current model explicit: how work is organized, how decisions are made, how information flows, how resources are allocated. Only after making the implicit explicit is it possible to assess whether the current model supports the intended strategy.

A useful tool is the exercise of mapping critical decisions: listing the 20-30 most important decisions for strategy execution (e.g., accepting or rejecting an order, setting prices for new clients, approving equipment investment, hiring key staff, launching new products), and for each asking: who decides, based on what information, using what criteria, within what timeframe, with what escalation mechanism if there is disagreement. In most SMEs, this exercise reveals a lack of clarity about who decides what, that critical information is unavailable, that criteria are implicit and inconsistent, and that decision timelines are excessive.

Second implication: accept that there is no optimal operating model—there are irreducible trade-offs. The temptation is to seek a model that maximizes both efficiency and flexibility, centralization and autonomy, control and trust. The reality is that these are genuine trade-offs: gain in one dimension implies loss in another. Conscious choice between trade-offs, aligned with competitive strategy, is what defines successful operating models.

A company competing on customization and customer proximity should not try to implement the operating model of a company competing on cost and scale. This seems obvious, but practice shows otherwise: growing SMEs often import practices from large companies (rigid processes, centralized decisions, heavy control systems) without recognizing that these practices were designed for a different competitive context.

The critical question is not "which operating model is better," but "which trade-offs does our strategy require us to make." A company choosing to compete on innovation must accept lower operational efficiency; a company choosing to compete on cost must accept less flexibility. Trying to have both results in an incoherent operating model that supports neither strategy effectively.

Third implication: invest in execution capability before increasing strategic ambition. Evidence shows that most strategic failures do not result from the wrong strategy, but from the inability to execute the chosen strategy. For SMEs with weak operating models, the priority is not to refine strategy—it is to build execution capability.

This has a counter-intuitive implication: it may be more profitable to invest in improving execution of the current strategy (even if it is not the optimal strategy) than to switch to a better strategy that the company cannot execute. An industrial SME with a quality differentiation strategy but inconsistent production processes gains more by investing in standardization and quality control than by pivoting to an innovation strategy requiring capabilities it does not have.

The correct sequence is: first, build an operating model that consistently executes the current strategy; second, use that reliable execution base to experiment with strategic extensions; third, only after validating that the new strategy is viable, redesign the operating model to support it. The wrong sequence—changing strategy without changing the operating model—is the most common cause of strategic failure in SMEs.

Fourth implication: recognize that operating model redesign is a transformation project, not a continuous improvement initiative. Incremental changes to operating model components (improving a process, implementing a system, reorganizing a department) are useful but not sufficient. Operating model redesign requires systemic change—multiple components changed in coordination, over 12-24 months, with significant management time investment.

The change management literature shows that successful transformations require five conditions: first, a clear case for change (why the current model does not support the strategy); second, a clear vision of the future model (how it will be different and why); third, a transition plan (sequence of changes, interdependencies, required resources); fourth, active resistance management (identifying affected stakeholders, anticipating objections, building a support coalition); fifth, progress monitoring and adjustment (transition metrics, frequent review, correction of deviations).

Most SMEs fail in one or more of these conditions. The case for change is unclear—"we will improve processes" is not specific enough. The vision of the future model is vague—"we will be more efficient" does not define operational architecture. The transition plan does not exist—changes are launched without coordination. Resistance is not managed—it is assumed that "it makes sense, so people will go along." Progress is not monitored—there are no transition metrics, only final outcome metrics.

Fifth implication: consider that operating model redesign may require external expertise. Evidence shows that successful operating model transformations often involve external support—not because companies lack intellectual capacity, but because they lack specific experience in organizational design, and because change requires an external perspective to challenge implicit assumptions.

The decision to use external support depends on three factors: first, redesign complexity (the more operating model components need to change simultaneously, the greater the benefit of external experience); second, availability of management time (operating model redesign consumes 20-30% of CEO and CFO time over 12-18 months—if that time is not available, the transformation does not happen); third, risk of failure (the higher the risk that transformation failure compromises company viability, the greater the benefit of reducing failure probability through external support).

The common mistake is to see consulting as a cost to minimize. The right question is not "how much does it cost," but "what is the value of increasing the probability of success from 40% to 70%." For a transformation whose success is worth €2M/year in incremental EBITDA, an investment of €100-200k in external support that increases the probability of success by 30 percentage points has a 3-6× return in the first year. Well-designed management consulting is not a cost—it is an investment with measurable return.

Where the Topic Is Fragile

Research on operating models has three important limitations that decision-makers should recognize.

First limitation: most empirical studies focus on medium-large companies in developed economies. Generalization to micro-enterprises (fewer than 10 employees) or emerging market contexts is questionable. Micro-enterprises may lack the minimum scale to implement certain operating model components (e.g., ERP systems, specialized support functions, layers of middle management). Markets with weak institutions may make certain models unviable (e.g., trust-based models in contexts with low contractual enforcement).

Second limitation: the causal relationship between operating model and performance is difficult to rigorously establish. Successful companies have resources to invest in sophisticated operating models—but it is unclear whether the operating model causes success or success enables investment in the operating model. The most rigorous longitudinal studies (tracking companies before and after operating model changes) are rare and focus on large-scale transformations, not incremental changes.

Third limitation: the literature on operating models is dominated by management consultancies, whose incentive is to demonstrate that operating model redesign creates value (because they sell redesign services). Independent studies are less common, and when they exist often show more modest results. The meta-analysis by Bromiley and Rau (2016) suggests that the average effect of operational improvement initiatives is about half that reported in consultancy case studies.

These limitations do not invalidate the central argument—that the operating model matters for performance—but suggest caution regarding the magnitude of expected effects and the generalizability of prescriptions. What works for a German multinational with 5,000 employees may not work for a Portuguese SME with 50 employees. Adapting generic frameworks to specific contexts requires judgment, not mechanical application.

Open Questions

Four questions remain insufficiently answered by current research and represent learning opportunities for companies willing to experiment.

First question: how to design operating models for companies undergoing digital transition? The literature on operating models was developed in a pre-digital context, assuming that processes are executed by people and that information systems support processes. The current reality is that processes are increasingly executed by systems (automation, AI), and the boundary between process and system is blurred. It is unclear which operating model architectures are appropriate for companies operating simultaneously in traditional (manual processes) and digital (automated processes) modes, or how to manage the transition between modes.

Second question: how to design operating models for companies operating in ecosystems? Traditional literature assumes the operating model is internal to the company—defining how the company organizes its own resources to deliver value. The growing reality is that companies operate in ecosystems where value is co-created with partners, suppliers, and clients. It is unclear how to design operating models that cross organizational boundaries, or how to manage governance and incentive alignment in multi-organizational contexts.

Third question: how to measure return on investment in operating model redesign? The literature shows correlation between operating model quality and performance, but does not provide a robust methodology for measuring ROI of specific redesign initiatives. This makes it difficult to justify investment and prioritize among alternatives. Process metrics (e.g., cycle time reduction, increase in first-pass rate) are insufficient because they do not link to financial impact. Financial metrics (e.g., EBITDA increase) are difficult to causally attribute to operating model changes versus other factors.

Fourth question: how to design operating models that balance short-term efficiency with long-term adaptability? The tension between optimizing for the current context versus preserving options for future contexts is theoretically recognized but poorly resolved in practice. Operating models optimized for efficiency are rigid; models designed for flexibility sacrifice efficiency. It is unclear whether there are architectures that allow both simultaneously, or if the trade-off is irreducible and the only solution is periodic redesign.

These questions have no generic answer—they require contextual experimentation. Companies that consciously address them, document choices and results, and share learning contribute to advancing collective knowledge on operating model design.

Sources

• Baldwin, C. Y., & Clark, K. B. (2000). Design Rules: The Power of Modularity. MIT Press.
• Beer, M., & Nohria, N. (2000). Cracking the code of change. Harvard Business Review, 78(3), 133-141.
• Birkinshaw, J., & Gibson, C. (2004). Building ambidexterity into an organization. MIT Sloan Management Review, 45(4), 47-55.
• Boston Consulting Group (2019). The Operating Model of the Future. BCG Henderson Institute.
• Bromiley, P., & Rau, D. (2016). Operations management and the resource based view: Another view. Journal of Operations Management, 41, 95-106.
• Deloitte (2017). Rewriting the rules for the digital age: 2017 Deloitte Global Human Capital Trends. Deloitte University Press.
• Hammer, M., & Champy, J. (1993). Reengineering the Corporation: A Manifesto for Business Revolution. Harper Business.
• IAPMEI (2021). Relatório de Execução Portugal 2020: Programas de Apoio à Competitividade. Agência para a Competitividade e Inovação.
• INE (2022). Empresas em Portugal 2020. Instituto Nacional de Estatística. https://www.ine.pt
• Kaplan, R. S., & Norton, D. P. (2004). Strategy Maps: Converting Intangible Assets into Tangible Outcomes. Harvard Business School Press.
• Kotter, J. P. (2012). Leading Change (2nd ed.). Harvard Business Review Press.
• Levitt, B., & March, J. G. (1988). Organizational learning. Annual Review of Sociology, 14, 319-340.
• McKinsey & Company (2018). Performance through people: Transforming human capital into competitive advantage. McKinsey Global Institute.
• Porter, M. E. (1985). Competitive Advantage: Creating and Sustaining Superior Performance. Free Press.
• PwC (2016). 2016 Digital IQ Survey: European findings. PricewaterhouseCoopers.
• Ross, J. W., Weill, P., & Robertson, D. (2008). Enterprise Architecture as Strategy: Creating a Foundation for Business Execution. Harvard Business School Press.
• Rousseau, D. M., Sitkin, S. B., Burt, R. S., & Camerer, C. (1998). Not so different after all: A cross-discipline view of trust. Academy of Management Review, 23(3), 393-404.
• Simons, R. (2005). Levers of Organization Design. Harvard Business Review, 83(3), 94-103.
• Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and strategic management. Strategic Management Journal, 18(7), 509-533.
• Tushman, M. L., & O'Reilly, C. A. (1996). Ambidextrous organizations: Managing evolutionary and revolutionary change. California Management Review, 38(4), 8-30.
• World Management Survey (2019). Management practices in Portugal: 2019 survey results. London School of Economics. https://worldmanagementsurvey.org
• Banco de Portugal (2021). Productivity dispersion and the role of firm-specific factors. Economic Bulletin, October 2021.
• AICEP (2022). Internacionalização da Economia Portuguesa: Relatório Anual 2021. Agência para o Investimento e Comércio Externo de Portugal.