This post synthesises the literature on what happens when Design enters organisations governed through the programme management tradition - the stage-gate reviews, programme boards, benefits realisation frameworks, and upward accountability structures that New Public Management reforms embedded across UK public services, and that remain the dominant governance culture in the NHS and much of the wider public sector. It draws together work I've been developing across several series: the epistemological tension between design and programme management, the performance-and-substance gap, the metaphorical fragmentation of design roles, and the psychodynamic costs of design practice in contested institutional contexts.
The diagnostic runs from Hill's (2012) widely-cited observation that designers are naive about organisational power, through to a psychodynamic account of what that naivety costs the designer personally. The implication isn't that design in programme cultures is hopeless - it's that entering such cultures with design intentions requires an honest reckoning with the dynamics at play before you start drawing journey maps.
The problem stated plainly
Design enters programme cultures in many forms, but the organisational terrain is recognisable. A large NHS trust restructuring its data infrastructure through a vendor-led delivery programme. A government department running a transformation portfolio managed by consultants from one of the big four firms, with workstreams governed through stage-gate reviews and benefits realisation frameworks. A regional health authority coordinating digital change across multiple trusts, each with its own governance, its own legacy systems, its own political sensitivities - the whole thing held together by a programme board, a senior responsible owner, and a delivery manager whose primary currency is risk and schedule. In each case, the organisation has chosen to treat a complex, cross-cutting challenge as a programme: something to be decomposed into workstreams, sequenced into phases, tracked against milestones, and governed through highlight reports and gateway reviews.
The designer arrives into these cultures and finds that the problems they are asked to address - fragmented user journeys, poor adoption, services that technically function but don't actually work for the people using them - are produced by the very governance structures within which they must now operate. This is not a communication problem or a matter of learning the right vocabulary; it is a structural condition. The literature gives it several names, but they converge on the same diagnosis: when organisations are organised around products, projects, or programmes, design work - which is inherently cross-cutting, relational, and systemic - meets institutional resistance not because anyone opposes it, but because the governance structures select against it.
The diagnostic: why programme cultures resist design
Naive about power
Hill (2012) frames this as the "dark matter" of organisational culture, policy environments, market mechanisms, legislation, and finance models shapes what services actually do, but sits beneath the surface of what design processes typically engage with. Designers, Hill argues, too often appear naive about how organisations actually make decisions. The user is rarely aware of the organisational context that produced a service, yet the outcome is directly affected by it. Hill's argument is that designers need to engage with this substrate - and that too few do.
But Hill's diagnosis - that designers are naive about organisational reality - is the mild version. A more uncomfortable body of work asks why design is systematically naive, and whether the naivety is a feature rather than a bug.
Design's idealist blindness
Von Busch and Palmås (2023), in The Corruption of Co-Design, argue that design operates through "idealist-tainted glasses" - a worldview in which the social world is treated as raw material waiting to be shaped by empathetic designers, and where the best argument wins on its merits. Their central charge is that design discourse lacks a language for competing interests and power dynamics. In a vivid passage describing a design school critique in Gothenburg, they observe that students present transport design proposals as if the car industry, the fossil fuel lobby, and the political interests that sustain car-centric planning simply don't exist. The entire room has "tacitly agreed on sidelining, or at least ignoring, the most vested and dollar-rich interests in the region".
This isn't a peripheral observation about student work. Von Busch and Palmås argue it reflects a structural feature of how design conceives of change: designers suffer from the conviction that the best idea wins, treating design as a game of persuasion rather than a contest of power. Their remedy - which they call "Realdesign" - asks designers to replace the idealist's "what if?" with the realist's "who whom?" - who benefits, who pays the price, who holds the power to make this real or block it? The core insight is this: visions that are not backed by a realistic perspective on power remain toothless, however well-researched and beautifully presented.
Their first proposition - "Design must get real" - cuts particularly close to the programme management context: design proposals get activated not through the quality of their argument but through tensions between networks of power, incentives, and influences. A journey map will not change a procurement framework. A service blueprint will not shift a governance model. What makes things happen is power, not persuasion.
The tools carry it
Tonkinwise (2010, 2023), cited by von Busch and Palmås, sharpens this into a specific methodological critique. In his 2010 Core77 essay, Tonkinwise argued that "being ethical, in order to avoid politics, is a political position" - designers doing social innovation unwittingly align with neoliberal governance by designing community replacements for state services.
His 2023 SVA lecture, "All Care, No Responsibility", extends this into a full critique of service design's tools. The service blueprint, he argues, is fundamentally a management tool - "a script by which a manager will discipline workers" unless you design it differently. The theatrical metaphor of front-stage and back-stage, the "line of visibility" conceals the labour that makes the service work, and in so doing makes it look magical by black-boxing the work. "The service worker is merely a box within the overall blueprint". This is, in Rancière's (2013) terms, a distribution of the sensible built into the tool itself: the blueprint determines what is visible and what is noise, whose experience counts as designable and whose is rendered invisible by the representational form.
Tonkinwise traces this to service design's origins in early-1980s neoliberalism: "Services become designable as the state stops caring, as the state is stopped from caring". The tools of service design - blueprints, personas, journey maps - systematically conceal what Hochschild (2012) identifies as the emotional labour of care work - controlling your own emotions, performing prescribed emotions, reading and managing customers' emotions.
And design's focus on the user experience misses what he calls "contexts of aggregation" - the systemic conditions, the historic struggles, the political settlements that produced the current arrangement. Empathy sees the individual interaction but not the system. It cannot see the silenced voices, the structural conditions, or the power relations that determine outcomes.
His counter-proposal: the way to address this isn't better empathy but designing the customer - using anthropological insights about gift-giving, commons governance, and graduated reciprocity to create genuine relationships rather than transactional encounters. And his closing line captures the honest limit: "Design is not the answer. Design can contribute to the creation of sociality, the restoring of sociality".
This is a direct challenge to how design is typically positioned in programme-management contexts. The standard pitch - "we bring the user perspective" - assumes that the user perspective is what's missing, and that surfacing it will cause change. Tonkinwise's point is that the user perspective may be perfectly well understood by the programme and simply overridden by other priorities. The bottleneck isn't informational; it's political.
The institutional logics produce it
Kimbell (2021) approaches the same problem from a different angle. In Logics of Social Design, she identifies three institutional logics through which design operates in social and public sector contexts: innovation-austerity, deliberation-pluralism, and anticipation-utopia. Each logic shapes what designers are permitted to do, how their legitimacy is established, and what they're structurally prevented from seeing.
The innovation-austerity logic is the most relevant to programme-led contexts. Under this logic, designers are positioned as enablers of innovation - responsive to social needs, facilitating co-production, amplifying local resources. But, Kimbell argues, this logic "downplays the political conditions in which social design is organised and carried out". Critical voices are tolerated but marginalised. Designers are advised to think of themselves as responsive rather than responsible for outcomes - which, Kimbell notes, "might be seen as being reactive and tactical, rather than active and strategic in the face of complex, and political issues".
The anticipation-utopia logic (design as visionary practice, proposing futures beyond current frames) has a different but related blind spot: it mutes "acknowledgement of the political and economic conditions that sustain such institutional forms". This is the logic that underpins Dorst's frame creation, Dunne and Raby's speculative futures, and Folkmann's ontology of the actual and the possible - all of which position design as reaching toward desirable futures that the present has not yet imagined. The designer as visionary positions themselves outside the system they're trying to change, which means they never have to reckon with the structural constraints that make the system the way it is.
The naivety Hill describes isn't accidental. It is produced by the institutional logics within which design operates - logics that, as Rancière (2013) would frame it, distribute what is sensible within the practice, determining which political dimensions of design work can be perceived and which are structurally invisible. Designers working within programme-led organisations are channelled toward the innovation-austerity logic - asked to be helpful, responsive, and user-focused, while the governance structures, funding models, and power relations that actually determine outcomes remain off-limits.
Design as already co-opted
Bailey (2021), as I discussed in Owning the Problem Space, takes this furthest: design in government contexts has been discursively and practically remodelled to align with existing institutional preoccupations. What is called "design" may already have been shaped to avoid the confrontations that genuine problem-space ownership would require; and if the design practice has already been co-opted to fit the governance logic it ought to challenge, the question of what design can actually reach becomes considerably more difficult.
How co-option works in practice
Each of the authors above describes a different mechanism through which design gets captured. Bailey (2021) argues that "design for government" has been discursively remodelled - reconstituted to align with a dominant political dogma about fixing the public sector through innovation. The apparatus is, as she puts it, "seductive rather than effective": it produces performances of change that enrol participants into believing they are engaged in innovation, while structural decisions have already been taken. The field grows not through evidence of results but through "continual recruitment of new enthusiasts", offering participants an identity upgrade - creative, experimental, innovative - that makes the apparatus self-sustaining.
Kimbell (2021) identifies a complementary dynamic through her institutional logics framework: the innovation-austerity logic positions designers as "responsive, rather than responsible for outcomes" - muting their engagement with the political conditions that shape those outcomes. Julier and Kimbell (2019) originally discussed this in relation to the concept of virtualism: design tools impose an image of expert knowledge that bargains for professional legitimacy, while the actual interventions remain "virtual, not actual" - performing innovation without delivering it.
Von Busch and Palmås (2023) approach the problem from a different angle entirely, tracing how the material protocols of co-design - Post-it notes, presentation formats, PowerPoint, planning templates - act as agents of betrayal, systematically filtering proposals into what can be visualised and processed within existing institutional formats. "The radicalism of the original visions was disarmed by design", they write of a Gothenburg co-design process; the betrayal wasn't deliberate but was built into the inscriptions through which ideas had to pass. Tonkinwise (2023), in his lecture on service design and care, argues that the tools themselves carry their origins: the blueprint as management script, the line of visibility as labour concealment, empathy as the substitution of individual attention for systemic analysis. Using these tools uncritically reproduces the neoliberal logic they were designed under.
The work in Bisset and Kuštrak Korper (2023) offers one additional lens on how this operates at the level of everyday practice. Studying service design roles in Swedish public sector contexts, we identified ten generative metaphors through which "the service designer" is understood: healer, plumber, teacher, mechanic, entertainer, political actor, bureaucrat, visualiser, humaniser, sensemaker. These aren't just different descriptions of the same role - they generate different expectations, different success criteria, and different practices. When a project manager expects a mechanic and gets an evangelist, or expects a bureaucrat and gets a political actor, frustration follows.
Most of these metaphors have little to do with actually designing services. A holist - the idea of the service designer as someone who can transcend organisational silos, or integrate disparate touchpoints through a unifying service journey - integrates existing fragments but doesn't design new wholes and usually lacks the political mandate to act on such insights. A teacher metaphor for the role of service designer - see service designers building "user-centred" capability but doesn't necessarily produce designed or tangible material outcomes. A visualiser makes things visible - but as I've argued elsewhere, visibility alone doesn't produce change. "If something called Service Design happened here, what purpose did it serve, and what in fact (if anything) did we design?"
Metaphorical fragmentation is one way design gets captured without anyone deciding to capture it: each metaphor serves institutional functions beyond understanding, and the performance of design (workshops, maps, canvases) substitutes for the substance of designed change - perhaps just allowing the decision-makers to cite a journey map or blueprint of the status-quo as evidence of now being user-centred. This operates alongside the other mechanisms described above - Bailey's discursive remodelling, Kimbell's institutional logics, Julier and Kimbell's virtualism, von Busch and Palmås's protocol betrayal - and together they produce something more than the sum of their parts.
If the metaphor for design itself is contested - if ten different stakeholders project ten different meanings onto "service design" - then the practice that's supposed to clarify meaning and 'integration' is itself mired in confusion. Design can't bootstrap shared understanding of other domains or contexts if there's no shared understanding of what design is. This is the practical, operational version of what Kimbell describes at the level of institutional logics: the role isn't just constrained by the institution - it's constituted differently by every stakeholder who encounters it.
The critiques converge on the same unsettling point: the question isn't just whether programme cultures resist design - it's whether the version of design that survives within those cultures has already been defanged. Von Busch and Palmås would say this is predictable: design's idealist self-image means it doesn't recognise when it has been captured, because it doesn't have a vocabulary for capture. It sees collaboration where a realist would see co-option.
Fayard, Stigliani and Bechky (2016) provide the occupational sociology that explains why this vulnerability is structural. Their study of how service designers construct their professional mandate found that the occupation differentiates itself primarily through ethos - holism, empathy, co-creation enacted through material practice - rather than through protected technical expertise. The tools can be learned by anyone; management consultants and marketers can adopt journey mapping, prototyping, and visualisation without difficulty. What makes these practices distinctively "design" is the values behind them. This is precisely what makes design so susceptible to institutional capture: when organisations adopt journey mapping without holism, persona work without genuine empathy, or workshops without real co-creation, they have acquired the form while discarding the substance that gave it meaning. The mandate that rests on ethos rather than jurisdictional closure is a mandate that can be hollowed out from within.
Each metaphor captures at most one of Fayard et al's three values - the holist offers integration without empathy, the humaniser offers empathy without systemic framing, the entertainer performs co-creation as workshop theatre - but none carries the integrated ethos. Fragment the role into separate metaphors and you fragment the ethos; what remains is the performative shell, each piece serving institutional functions while the integrated practice that might challenge the programme logic has been quietly disassembled.
What happens to the designer: social defences and emotional labour
In my other NORDES paper (Bisset, 2023b), I asked the question: what happens to the designer caught inside these dynamics? Drawing on psychodynamic theory (Fotaki and Hyde's (2014) organisational blind spots, work from the Tavistock tradition, Wastell's (1999) social defences in information systems development), the paper identifies a specific pattern: designers in public sector contexts are simultaneously idealised - expected to work miracles, to be "Service Design Jesus" - and expected to absorb the negative emotions their inquiries provoke. Organisations project both their hopes and their anxieties and fragmented understanding and representations of what design is and what it can do onto the designer, then blame the designer when the structural constraints the organisation refuses to confront, or as Tonkinwise, Bailey and Kimbell suggest design is incapable of confronting remain unchanged.
This is the emotional labour argument that Tonkinwise (2023) makes about service workers, applied to the designer themselves. The splitting is precise: the ideals of the project (transformation, participation, collaboration) are separated from what it takes to realise those ideals (confronting power, surfacing uncomfortable truths, managing resistance). The designer is expected to carry both sides - to be the visionary and the lightning rod - while the organisation avoids doing the difficult work of implementing change. When the project doesn't deliver miracles, it's easier to project blame onto the designer, or as Kimbell and others suggest marginalise them, or carry on the empty performance, than to reflect on or transform the structural conditions that made the miracles impossible, or led to such desperate idealism in the first place.
The paper identifies a blank space in the design literature: despite all the talk of design as socially transformative practice, there are almost no tools or frameworks for helping designers handle the psychodynamic impacts of this work - the idealisation, projection, splitting, and blame that are predictable features of design practice in contested institutional contexts. The Systems Engagements series develops this psychodynamic apparatus in more detail - from Bion's basic assumptions in co-creation workshops to the BART framework and countertransference as the designer's primary instrument for reading what the system is projecting. Design research needs, as the paper argues, to help practitioners "depersonalise the impact of our design work upon ourselves as careworkers".
Without the psychodynamic layer, the other critiques remain intellectual exercises. The "Service Design Jesus" pattern is what makes the critique felt - it's the lived experience of what happens when a designer enters a programme-led culture with even humble materialist or envisionary intentions and meets all the dynamics the literature describes.
NPM governance selects against design
Boztepe (2023) grounds this empirically. Studying design capability building in Swedish municipalities, she identifies "tissue incompatibility" - design's ways of working meeting institutional logics they cannot easily integrate with. Crucially, the incompatibility goes both ways: the public sector mindset doesn't readily embrace design, but designers also struggle to adapt their methods to institutional constraints. The three tensions she documents - between insider/outsider positioning, incremental vs transformative ambitions, and the difficulty of demonstrating impact - are recognisable to anyone who has worked as a designer embedded in a public sector programme.
The structural argument, which I've developed more fully in Beyond Technomagic, draws on amongst other perspectives the military design literature and New Public Management (NPM) critique: NPM governance structures don't merely fail to support design - they actively select against it. Business cases require quantified benefits before the design work that would identify real benefits has been done. Stage gates demand deliverables at fixed intervals, but design produces understanding rather than deliverables. The planning/design confusion means design work gets governed with planning's expectations, timelines, and success criteria applied to an activity they don't fit.
Boland and Collopy (2004) provide the management theory that explains why this selection operates at the level of organisational cognition, not just governance procedure. Their distinction between a "decision attitude" and a "design attitude" in management identifies two fundamentally different orientations to problems. The decision attitude - which Boland and Collopy argue is "overwhelmingly dominant in management practice and education today" - treats problem-solving as making rational choices among pre-defined alternatives, using tools of economic analysis, risk assessment, and calculation. The design attitude, by contrast, "views each project as an opportunity for invention that includes a questioning of basic assumptions". The evaluative mechanism in the design attitude is what Boland and Collopy call a "sense of fit" - the judgement of whether materials, technologies, logics, objectives, and scope work together in harmony - and they insist that this sense of fit, "like an aesthetic judgement, is a subjective matter and should not be relinquished to a technique of calculation".
Programme governance is the decision attitude made institutional. RAG ratings, benefits realisation frameworks, and gateway reviews all operate by comparing current state against pre-specified criteria - which is precisely the evaluative logic Boland and Collopy argue is antagonistic to design. The language of decision and increase, they write, "is inherently antagonistic to the language of design and balance". When a programme board asks whether a design workstream is on track, it is asking a decision-attitude question - does the output match the specification? - when what design requires is a continuous evaluative engagement with whether the emerging work coheres, whether it serves its purpose, whether the composition holds together.
Stage gates compress this continuous evaluative capacity into discrete checkpoints that cannot bear the weight placed on them; the governance review asks "has the process been followed?" when the design question is whether the process has produced something worth following. Strathern (2003) identifies this as a general feature of audit cultures: accountability is discharged by demonstrating the existence of control systems, not by demonstrating the quality of what those systems govern. The programme board's RAG rating assures the quality of the governance apparatus; it says nothing about whether the service being governed will work for the people who use it. The decision attitude and the audit culture are, in this sense, mutually reinforcing: audit provides the institutional mechanism, while the decision attitude provides the cognitive orientation that makes checking processes against criteria feel like adequate evaluation.
And as I argued in Performance and Substance, design itself can function as a "rationalised myth" - adopted for legitimacy rather than effectiveness. A project can have all the trappings of design (workshops, post-it notes, journey maps, prototypes) while producing little genuine change. The design performance substitutes for substantive transformation. Boland and Collopy's framework reveals why this substitution is so easy to accomplish: once the decision attitude dominates, the organisation has already lost the evaluative capacity that would allow it to distinguish between a design performance and a designed outcome. The sense of fit has been replaced by the checklist; and the checklist, by definition, cannot tell you whether what you have made is any good.
From diagnosis to practice
The implication of this layered diagnostic isn't despair; it's that entering a programme-led culture with design intentions requires an honest reckoning with these dynamics before you start drawing journey maps. The question isn't "how do I introduce design thinking?" - it's "what version of design can survive and be productive in this institutional logic, and where does it need to push beyond what the institution is comfortable with?"
Programme cultures are not monolithic. As I explored through the lens of promise theory and accountability, programmes oriented toward risk, efficiency, provider management, or demand sensing create different conditions for design - different blind spots, different failure modes, and different versions of co-option. The people aren't hostile to design; they've simply never had reason to think in service terms. But the governance structures - multi-year funding cycles, programme-level accountability, product-organised teams - actively reproduce the fragmentation that a service lens would challenge. You're not walking into opposition; you're walking into a structure that doesn't have the connective tissue design needs in order to work.
Malmberg (2017) provides a useful bridge here. Her absorptive capacity model identifies three sequential phases through which design capability develops in organisations: acquisition (individuals encounter design and form initial understanding), assimilation (the organisation diffuses design knowledge and builds enabling structures), and exploitation (design generates value through sustained practice). The critical finding is that knowledge acquisition without the organisational capacity for assimilation leaves things functionally unchanged - designers are hired, methods are demonstrated, but the enabling structures are never built.
The gap between acquisition and assimilation is precisely where the dynamics described above do their work: design performances substitute for designed outcomes, journey maps get filed rather than acted upon, and the decision attitude continues to dominate unchallenged. The test isn't whether you produce design artefacts; it's whether those artefacts change decisions. Von Busch's "who whom?" question applies directly: who holds the power to act on what your maps reveal, and what would it take to mobilise that power? If the answer is "nobody in this governance structure", then the design work needs to be directed at the governance structure itself - which is Hill's dark matter, and Kimbell's point about engaging with political conditions rather than just being responsive within them.
Being explicit about role expectations from the outset is one practical way of resisting the metaphorical fragmentation that allows design to be absorbed. A hiring manager may see cross-cutting sense-making; programme managers may see another workstream producing deliverables; external partners may see someone who makes onboarding easier. When "service design" means something different to every stakeholder, the practice that's supposed to clarify meaning is itself mired in confusion; naming which of these roles you are playing - and which you are not - is a way of preventing that absorption before it begins.
What practical strategies exist for building the assimilation infrastructure that Malmberg describes - for working with the grain of programme governance rather than against it - is the subject of the earlier post in this series that addresses practice directly. What this post contributes is the diagnostic framework that makes those practical strategies legible: without understanding why programme cultures select against design, the practical responses risk becoming another form of naive optimism.
References
- Bailey, J.A. (2021). Governmentality and power in 'design for government' in the UK, 2008-2017: An ethnography of an emerging field [Doctoral dissertation, University of Brighton].
- Bisset, F. (2023b). Crucifying 'Service Design Jesus': Exploring social defences encountered in the design process [Conference paper]. Nordes 2023.
- Bisset, F., & Kuštrak Korper, A. (2023). What did we actually design here and what purpose did it serve? Some generative metaphors for understanding service design in the Swedish public sector [Conference paper]. Nordes 2023.
- Boland, R., & Collopy, F. (2004). Managing as designing. Stanford University Press.
- Boztepe, S., Linde, P., & Smedberg, A. (2023). Design making its way to the city hall: Tensions in design capacity building in the public sector [Conference paper]. IASDR 2023. https://doi.org/10.21606/iasdr.2023.458
- Downe, L. (2015, June 22). Good services are verbs, bad services are nouns. GDS Design Notes.
- Fayard, A-L., Stigliani, I., & Bechky, B.A. (2016). How nascent occupations construct a mandate: The case of service designers' ethos. Administrative Science Quarterly, 62(2), 270-303.
- Fotaki, M., & Hyde, P. (2014). Organizational blind spots: Splitting, blame and idealization in the National Health Service. Human Relations, 68(3), 441-462.
- Hill, D. (2012). Dark matter and Trojan horses: A strategic design vocabulary.
- Hochschild, A.R. (2012). The managed heart: Commercialization of human feeling.
- Julier, G., & Kimbell, L. (2019). Keeping the system going: Social design and the reproduction of inequalities in neoliberal times. Design Issues, 35(4).
- Kimbell, L. (2021). Logics of social design [Conference paper]. Design as Common Good Online Conference.
- Malmberg, L. (2017). Building design capability in the public sector: Expanding the horizons of development [Doctoral dissertation, Linköpings University].
- Rancière, J. (2013). The Politics of Aesthetics. Bloomsbury Publishing.
- Strathern, M. (2003). Audit Cultures: Anthropological Studies in Accountability, Ethics and the Academy. London: Routledge.
- Tonkinwise, C. (2010, September 1). Politics please, we're social designers. Core77. https://www.core77.com/posts/17284/
- Tonkinwise, C. (2023). All care, no responsibility: Everything service designers need to know about politics but were afraid to ask [Guest lecture]. MFA Products of Design, School of Visual Arts.
- von Busch, O., & Palmås, K. (2023). The corruption of co-design.
- Wastell, D.G. (1999). Learning dysfunctions in information systems development: Overcoming the social defenses with transitional objects. MIS Quarterly, 23(4).