ICR Research 2026 #7: When Process Defeats Purpose
Procedural Complexity, Workforce Capability, and the Case for Operating Model Reform
The United Kingdom’s Soft Power Council launched on 15 January 2025, co-chaired by the Foreign Secretary and the Secretary of State for Culture, Media and Sport. By July 2025 it had reviewed a draft strategic framework with proposed goals and outcomes. At the time of writing, the strategy has not been published.¹ This prompts a question our comparative research raises directly: why do institutions with real strategic intent and significant soft power assets so often struggle to convert them into measurable outcomes? The evidence suggests the issue may lie with their operating models.
This article addresses two of the four elements of the soft power operating model introduced in the second piece in this series: procedures and governance on one hand, and workforce and capability on the other. The comparative evidence suggests these two elements are connected at a deeper level than is usually acknowledged. The procedural complexity that slows institutions and the career structures that shape workforce incentives are both manifestations of the same underlying challenge: operating models designed for a different era, maintained by institutions whose accountability requirements have grown faster than their underlying architectures have been updated.
What the Reform Literature Tells Us
Three bodies of work that are rarely brought into dialogue with soft power research provide the analytical foundation for what follows.
Jennifer Pahlka, writing from her experience leading the United States Digital Service and Code for America, argues in Recoding America that government underperformance is not primarily a problem of resources or intentions but of operating systems: the accumulated rules, procedures, and processes through which policy intentions must pass before they can reach their intended beneficiaries.² Her central finding is that procedural accretion is the predictable result of rational responses to specific problems, each layer justifiable in isolation, collectively producing systems that resist the delivery of their stated purposes. Pahlka developed this analysis in the context of domestic digital service delivery in the United States federal government, and the institutional setting of international cultural relations differs in important respects. But the underlying dynamic she identifies, in which procedural requirements accumulate faster than they are reviewed, and accountability systems designed for one purpose are extended to others where they fit poorly, is not specific to that context. It is a recognisable feature of public institutions under sustained accountability pressure, and it offers a productive analytical lens for examining why some soft power systems underperform despite adequate resourcing.
Beth Simone Noveck, Director of the Governance Lab at New York University, identifies in Solving Public Problems a structural gap between the capabilities public institutions require and those that conventional career paths produce.³ Her argument centres on what she terms the public problem solver: a practitioner combining analytical skill, collaborative design capability, and the ability to work across institutional boundaries. The skills she identifies as missing from conventional public administration, specifically data literacy, collaborative problem-definition, and iterative programme design, are recognisable to practitioners in cultural diplomacy and international education as precisely those that their fields require and that standard civil service formation does not reliably develop.
Geoff Mulgan, Professor of Collective Intelligence, Public Policy and Social Innovation at UCL, contributes two arguments of relevance. In Big Mind, he contends that the central challenge facing complex organisations is not the quality of individual decisions but the quality of the systems through which information, learning, and decision-making are connected.⁴ For soft power, this reframes the question: whether individual programmes succeed or fail matters less than whether the systems running them can learn and adjust accordingly. More recently, his work on the next generation of public institutions has argued that most public organisations still operate on architectures that predate digital transformation and the current pace of geopolitical change, and that the design challenge is to move from periodic reform cycles to continuous adaptive capacity.⁵
What the Comparative Research Shows
Our comparative research for the British Council assessed soft power performance across twenty-five jurisdictions using the AIO framework, which distinguishes between assets, institutional infrastructure, and outcomes. Its central finding is that strategic coherence, the alignment of policy objectives, institutional mandates, and programme delivery, is a stronger predictor of outcome performance than resource scale.⁶ Countries achieving superior efficiency ratios are not uniformly those with the largest budgets. Germany achieves the highest absolute outcome score through an institutional architecture that maintains clear divisional mandates across the Goethe-Institut, DAAD, and GIZ, with each organisation building specialist capability over decades within its defined domain. South Korea achieves the highest efficiency ratio by concentrating resources around a clearly identified comparative advantage in cultural production.⁷
What the research does not directly measure is the internal governance processes of the institutions concerned: their approval chains, risk frameworks, or workforce development structures. The efficiency differentials it documents are consistent with Pahlka’s hypothesis about what causes institutional underperformance, but the research does not independently confirm that hypothesis at the level of specific procedural mechanisms. The connection between the comparative outcome data and the reform literature is therefore analytical rather than directly evidential and should be read as such.
What the research does directly support is the importance of feedback architecture. The AIO framework was designed to provide exactly this: a mechanism for connecting evidence about outcomes back to decisions about assets and infrastructure, so that the system can adjust in response to what it learns. Mulgan’s argument that learning capacity must be built into institutional architecture rather than added as an evaluation supplement after the fact applies with particular force to soft power, where the outcomes of interest are slow-moving and relational, and where the distance between programme delivery and measurable impact is routinely long enough to defeat standard accountability cycles.⁸
The Measurement Problem
Procedural complexity in soft power is compounded by a measurement challenge with no close equivalent in most other domains of public expenditure. The outcomes of cultural and educational engagement, trust, relationship, and shifted perception, are inherently difficult to quantify in ways that align with standard accountability frameworks. Mulgan’s analysis of collective intelligence systems identifies this as a general challenge for organisations whose value is relational rather than transactional: when accountability systems are designed to measure transactions and the outcomes of interest are the gradual development of trust and understanding, the systems and the outcomes are structurally misaligned.⁹ The result, as he documents, is a systematic tendency for organisations to produce evidence of process compliance rather than evidence of impact, not from indifference to impact, but because process compliance is what their accountability frameworks reward.
This dynamic is directly relevant to the question of why strategic intent does not reliably convert into measurable outcome. Noveck makes the workforce implication explicit: accountability frameworks that reward process compliance develop and select for certain capabilities, specifically those associated with documentation and procedural adherence, at the expense of the collaborative and analytical skills that effective programme delivery requires.¹⁰
Three Principles Grounded in the Evidence
The comparative research, read alongside these analytical frameworks, points to three principles warranting attention in any serious operating model reform.
The first is mandate clarity. The comparative research finds that higher-performing systems consistently maintain clear institutional mandates with limited overlap. Germany’s model is the clearest example: the separation of mandates between cultural promotion, academic exchange, and development cooperation reduces the coordination costs that arise when responsibilities are shared, and allows each organisation to build specialist capability over time within a defined domain.¹¹ Pahlka describes the internal procedures that organisations develop to manage overlapping boundaries with each other as interface management process, a form of institutional overhead that accumulates without accountability for whether it serves either organisation’s purpose.¹²
The second is proportionality in governance requirements. Pahlka documents in detail how risk frameworks and approval chains migrate from domains where they were developed to domains where the risk profile is materially different, without assessment of whether the migration is appropriate.¹³ The comparative research does not directly test this claim, but the efficiency differentials it finds between systems with different governance architectures are consistent with it.
The third is systematic feedback architecture. The AIO framework was designed to provide the structural basis for this: clear data points at each stage of the assets-to-outcomes chain, enabling practitioners to identify where investments are and are not generating returns. Mulgan argues, and the comparative evidence supports, that learning capacity at this level must be embedded in institutional design rather than treated as an add-on evaluation function.¹⁴
Implications
A national soft power strategy grounded in this body of evidence would address four questions:
It would specify mandate boundaries between principal institutions in ways that reduce overlap.
It would propose a proportionality review of governance and compliance requirements.
It would set out a workforce development programme specifying the skills profile required and the career structures through which it will be developed.
And it would establish feedback architecture through which evidence about what works is systematically collected and used.
Each of these is documented in the comparative and reform literature as a characteristic of higher-performing systems. But the case for addressing them now is not primarily analytical. It is contextual.
As the earlier articles in this series have argued, the current international moment presents a specific and time-limited opportunity. The simultaneous retrenchment of the United States across its public diplomacy infrastructure, the contraction of China’s Confucius Institute network, and the erosion of the multilateral frameworks within which soft power has conventionally operated have created a structural opening for middle powers with credible institutions and strategic coherence.¹⁵ The UK possesses the assets to act in that space. Whether it can do so depends on whether its institutions can convert assets into outcomes at the speed and scale the moment requires.
To do so requires rigour in designing the operating model. That rigour matters particularly for soft power because of what distinguishes it from most other instruments of foreign policy. Hard power effects are visible and relatively fast. Soft power effects are slow, cumulative, relational, and easily disrupted. A scholarship cohort whose experience of the UK is shaped by administrative friction rather than genuine engagement does not simply fail to generate influence: it generates the opposite. An alumni network that receives no sustained attention after the initial investment depreciates rather than compounds. A cultural partnership abandoned mid-cycle because an approval chain moved too slowly leaves a reputational residue that the next programme must work against. These are not hypothetical costs. They are the predictable consequences of operating models that were not designed for the specific character of soft power delivery: its long time-horizons, its dependence on genuine relationship rather than transactional exchange, and its acute sensitivity to the gap between what institutions promise and what they can actually deliver. Getting the operating model right is not a precondition for strategy. It is, in the current environment, part of what strategy means.
This is the seventh article in the ‘Recoding Soft Power’ series.
¹ Pannell, J. (2026) ‘Soft power is the edge in a hard power era’, British Foreign Policy Group, 6 May 2026; House of Lords Library (2026) ‘UK development policy: Soft power and global influence’, 23 January 2026.
² Pahlka, J. (2023) Recoding America: Why Government Is Failing in the Digital Age and How We Can Do Better. New York: Metropolitan Books.
³ Noveck, B.S. (2021) Solving Public Problems: A Practical Guide to Fix Our Government and Change Our World. New Haven: Yale University Press.
⁴ Mulgan, G. (2017) Big Mind: How Collective Intelligence Can Change Our World. Princeton: Princeton University Press.
⁵ Mulgan, G. (2025) ‘Building the Next Generation of Public Institutions’, Project Syndicate, 25 February 2025.
⁶ MacDonald, S. and Murray, A. (2026) Trends in Soft Power 2020–2025. London: British Council/ICR Research. https://doi.org/10.57884/FS51-KF14
⁷ Ibid.
⁸ Mulgan, G. (2017), op. cit., ch. 5; MacDonald, S. and Murray, A. (2026), op. cit.
⁹ Mulgan, G. (2017), op. cit., ch. 5.
¹⁰ Noveck, B.S. (2021), op. cit., pp. 47–73.
¹¹ MacDonald, S. and Murray, A. (2026), op. cit.
¹² Pahlka, J. (2023), op. cit., pp. 134–159.
¹³ Ibid.
¹⁴ Mulgan, G. (2017), op. cit., pp. 201–218; MacDonald, S. and Murray, A. (2026), op. cit.