'Design Is Not Craft' — Industry x Academia Roundtable

Proposition

“Design is not craft.” Is this claim correct, incorrect, or incomplete?

A record of simultaneous R1-R3 discussion among three industry personas (business-only) and three academic critics (critique-protocol compliant).

Related: design-craft-relationship-industry / design-pricing-vs-ai-commoditization (the billability of taste and the AI resilience of pricing models)

R1: Independent Arguments

Industry Personas

F (Design Firm, IDEO/frog type)

Position: We earn through billable hours on HCD processes and deliverable production. If craft is declared unnecessary, our revenue disappears.

Conclusion: Incomplete. Craft alone cannot sustain unit pricing, but abandoning it eliminates the basis for billing.

Hermes’s 40.5% operating profit margin (URD 2024) is a real case of craft investment, but structurally different from a firm. Hermes owns the product and sells it themselves. Firms create for clients, deliver, and transfer ownership. We are not positioned to pass craft premiums through to pricing.
Even if AI reduces production labor by 30% [to be confirmed], under human-hour billing this means revenue reduction. Erickson (2024)‘s “invisibilization of human contribution” means that time spent on prompt tuning and quality judgment is invisible to clients.
IDEO: Revenue declined from $300M to under $100M, headcount reduced approximately 50% over 3 years [to be confirmed]. Pivoted to capability building, but repeat engagements dried up (if you teach someone to fish, they fish for themselves the next year).
Conclusion to reject: “Abandon craft, shift to strategy/capability, and the firm will be fine.”

C (Strategy Consulting, McKinsey/Accenture type)

Position: We explain design ROI to C-suite and bill it as transformation engagements. We use proprietary research like MDI in proposals.

Conclusion: Correct as a sales pitch, but incomplete as a business fact.

When using MDI (top quartile +32pp revenue, +56pp total shareholder return over 5 years), we never mention “craft.” We sell it as “integrated management of customer experience.” However, this is correlation, not causation (SC-01, 2018, 300+ companies).
We raise Hermes’s 40.5% as a counter-evidence against our own position. Since a real case exists where craft investment is compatible with high margins, “craft is unnecessary” is factually incomplete.
In the AI era, relinquishing production turns us into a commodity of “just delivering strategy.” Only 16% of enterprises have completed business redesign per Deloitte survey [to be confirmed].
Nick Law’s “no one is talking about taste” is extremely difficult to productize. Stack Overflow Blog (2026-05): “What enterprises want is not the artisan slowly producing the finest quality, but someone who assembles hundreds of IKEA chairs in minutes.”
Conclusion to reject: “Everyone becomes a strategist.”

O (Independent Design Office, Pentagram/boutique type)

Position: Clients pay for “our name.” We earn through the premium of authorship.

Conclusion: Correct as a proposition, but incomplete as a business.

What clients pay for is judgment (what to make, what to discard), not the meticulousness of handwork. You cannot write “expensive because hand-drawn” on an invoice.
Yet the quality of judgment is formed through craft. Releasing the hand degrades judgment capability, and named commissions stop coming. Ive: “Material understanding can only be gained through hands-on work” (Dezeen 2016).
Craft serves three roles: (1) apparatus for forming judgment capability (2) apparatus for justifying premiums (3) signal for differentiation.
Scher (Pentagram) used Midjourney for performance.gov because “there was no budget” (D-01). Even at the Pentagram level, craft is compromised by budget constraints.
Conclusion to reject: “Preserving craft alone ensures you can earn.” Japanese traditional craft practitioners halved (114,000 to 54,000); Etsy sellers 8.1M to 5.6M.

R2: Cross-Rebuttals

Among Industry Personas

F to C: When you sell on ROI, clients go in-house the next year

What is the Year 2 engagement unit price for MDI-acquired deals? The better the ROI numbers, the more clients say “we can do this ourselves now” and move in-house. Deals won with ROI as a weapon kill repeat business proportional to their success. Show us the unit prices for deals where value-based pricing was accepted. Procurement departments push back: “Quote based on last time’s human-hour actuals.”

F to O: The three roles of craft do not become line items on a proposal

Listing “apparatus for forming judgment,” “apparatus for justifying premiums,” and “signal for differentiation” is meaningless when you cannot bill for them: unbillable things do not exist in business. Scher herself used Midjourney for performance.gov, with the reason being budget constraints. Even Pentagram compromises. The small-team model faces existential risk from a single AI-generated content lawsuit.

C to F: Holding craft has not prevented the billing basis from disappearing

IDEO’s contraction ($300M to $100M) occurred despite having craft. The problem is that “billable hours physically shrink.” As long as you stop at “it is serious because the contribution is invisible” without demonstrating how to make invisible contributions visible and billable, unit price decline will not stop.

C to O: Hermes’s profit margin is not a design office’s profit margin

What Hermes is paid for is the “Hermes” brand, not designers’ judgment capability. Whether judgment quality grows from craft experience or business experience is unverified. Pentagram has approximately 25 partners. This is a story about a few dozen people worldwide and does not constitute a survival strategy for the entire industry’s designers.

O to F: You cannot sell taste in a 400-person firm

Taste is tied to individuals. No client pays for “the taste of that 400-person firm.” If the firm genuinely wants to sell taste, it must downsize, but then it is no longer a “firm.” IDEO’s contraction is not a pivot but the result of being too large to pivot.

O to C: You contradict yourself by stripping craft from your own organization while calling it necessary

McKinsey Design’s “product builder” is a human who orchestrates AI agents --- that is, a human who has handed craft to AI. Saying “without craft you become a commodity” while outsourcing craft is a contradiction. The ROI model structurally cannot capture taste-driven engagements.

Post-R2 Common Recognition among Three Industry Personas

All three lack track records of operating post-AI billing models at a profit. Each is narrating a story convenient to their own position without having answered “what will generate revenue.”

R3: Conditions under Which Our Own Claims Would Be Wrong

Industry-Side Falsification Conditions

F1 (Taste premium is wrong): If AI-generated design is indistinguishable from human CDs in blind evaluation by CMO/CPO-level judges. Or if unit prices for named commissions contract relative to non-named AI-assisted engagements for two consecutive years.

F2 (Human-hour model collapse is wrong): If a major firm (200+ staff) fully deploys AI and maintains human-hour unit pricing with utilization rates above 80% for two consecutive years. Or if Fortune 500 RFPs never say “lower your rate if you use AI.”

F3 (Impossibility of practitioner reproduction is wrong): If alumni of firms that adopted AI as the primary means of OJT can independently lead senior CD engagements within 3 years (confirmed across 3+ firms). Or if a career path where juniors build AI-native portfolios without going through junior roles and receive direct named commissions from clients is established at a scale of 50+ cases per year.

Common structure among the three industry personas: All three depend on the “absence of data at this point.” Whether reading the absence of data as “therefore dangerous” or “therefore still unknown” reverses the conclusion.

R4: Synthesis --- Divergence Table

Points of Agreement (All 6 Participants)

“Design is not craft” is incomplete. All six participants refused to adopt this proposition at face value.
The speed at which AI substitutes craft production processes is accelerating, and no one can accurately predict the consequences. The absence of data is the shared premise.
The reproduction pathway for practitioners (next-generation designers/artisans) is threatened. The industry side identifies “disappearance of junior roles”; the academic side identifies “bypassing threshold concepts” and “reproduction of banking education” --- the same structure articulated in different vocabularies.

Divergence Table

Issue	Industry Side	Academic Side	Reason for Non-Convergence
Billability of taste/craft	All three acknowledge “no track record data of billing taste as a proposal line item that was accepted.” But F says “we must invent a billing model,” O says “it works only through individual-name commissions,” and C says “untranslatable into management language.”	The theoretical value of something and its conversion into market compensation are distinct questions. Hermes represents “enclosure of scarcity” not “realization of craft’s value.”	Value’s existence and value’s monetization are separate problems, and no theory bridges them.
Indispensability of embodied repetition	”Not moving the hand dulls judgment” (O) is an intuition without demonstrated causation.	Dewey’s continuity of experience, Schon’s reflection-in-action, and Polanyi’s tacit knowledge all support embodied repetition. But the scope of “embodied” is indeterminate (does digital manipulation count?).	The definition of “embodied” varies by author, and empirical verification is absent.
Survival of the human-hour billing model	F: Breaking (IDEO). C: Extending life through product builders but with revenue decline. O: Small team so margin improvement is possible.	Abbott’s jurisdiction theory: relinquishing production contracts jurisdiction and bargaining power. Braverman: logic of conception/execution separation.	Industry asks “how to earn,” academia asks “whose power is diminished” --- same phenomenon, different frames.
Consequences of AI democratization	The “efficiency dividend” problem: whether labor savings accrue to the client or the firm.	Re-stratification under the name of “democratization.” Pulling up the ladder from entry-level. Freirean reproduction of “banking education.”	Industry sees a zero-sum distribution problem; academia sees structural exclusion.

Items to Confirm (Accumulated across All Rounds)

Item	Content to Confirm
IDEO revenue contraction	$300M to under $100M (private company, based on former employee testimony)
MDI causality	+32pp/+56pp is correlation. Whether causation studies exist
Production labor reduction rate after AI adoption	Third-party measurements, not vendor/in-house announcements
Taste/CD fee proposal acceptance track record	Whether cross-industry quantitative data exists
Deloitte business redesign 16%	Identify methodology (sample, period, definition)
Pye’s continuum description	Where exactly in The Nature and Art of Workmanship (1968) the continuum is stated
Epistemological distance between Schon’s R-in-A and “craft”	Whether Schon’s (1983) architecture studio case (pp. 76-104) is isomorphic with material craft
Formation of pre-reflective evaluative ground in AI environments	Whether longitudinal studies exist

Practical Data to Confirm Next

Cases where taste/curation fees were established as independent billing line items after AI adoption
Trajectory of human-hour unit pricing and utilization rates at major firms after full AI deployment (2+ year period)
Number of actual cases of career paths where AI-native portfolios led to named commissions without going through junior roles

References

Key Data Cited by Industry Personas

Hermes URD 2024: Operating profit margin 40.5%, approximately 7,000 artisans, 76% in-house workshops
IDEO: Revenue $300M to under $100M, headcount approximately -50% [to be confirmed]
MDI (McKinsey 2018): Top quartile +32pp revenue, +56pp total shareholder return (correlation, 300+ companies)
Bain/Altagamma 2025: Luxury EBIT 2012 23% to 2025 outlook 15-16%
Stack Overflow Blog 2026-05-28: Artisans and Builders
Scher (Pentagram): Used Midjourney for performance.gov (budget constraint)
Japanese traditional crafts: Practitioners 114,000 to 54,000, production value 1/5 of peak
Etsy: Sellers 8.1M to 5.6M

Key Academic Theories Cited by Academic Critics

Simon, H.A. 1969/1996. The Sciences of the Artificial. 3rd ed. MIT Press.
Schon, D.A. 1983. The Reflective Practitioner. Basic Books.
Schon, D.A. 1987. Educating the Reflective Practitioner. Jossey-Bass.
Cross, N. 1982. “Designerly Ways of Knowing.” Design Studies, 3(4).
Dorst, K. & Cross, N. 2001. “Creativity in the design process.” Design Studies, 22(5).
Buchanan, R. 1992. “Wicked Problems in Design Thinking.” Design Issues, 8(2).
Pye, D. 1968. The Nature and Art of Workmanship. Cambridge University Press.
Ingold, T. 2013. Making. Routledge.
Sennett, R. 2008. The Craftsman. Yale University Press.
Polanyi, M. 1966. The Tacit Dimension. Doubleday.
Pallasmaa, J. 2009. The Thinking Hand. Wiley.
Adamson, G. 2013. The Invention of Craft. Bloomsbury.
Costanza-Chock, S. 2020. Design Justice. MIT Press.
Kimbell, L. 2011. “Rethinking Design Thinking.” Design and Culture, 3(3).
Abbott, A. 1988. The System of Professions. University of Chicago Press.
Braverman, H. 1974. Labor and Monopoly Capital. Monthly Review Press.
Dewey, J. 1938. Experience and Education. Kappa Delta Pi.
Papert, S. 1980. Mindstorms. Basic Books.
Lave, J. & Wenger, E. 1991. Situated Learning. Cambridge University Press.
Freire, P. 1970. Pedagogy of the Oppressed. Herder & Herder.
Vygotsky, L.S. 1978. Mind in Society. Harvard University Press.
Meyer, J.H.F. & Land, R. 2003. “Threshold Concepts and Troublesome Knowledge.”
Kirschner, P.A., Sweller, J. & Clark, R.E. 2006. “Why Minimal Guidance During Instruction Does Not Work.” Educational Psychologist, 41(2).
Ericsson, K.A., Krampe, R.T. & Tesch-Romer, C. 1993. “The Role of Deliberate Practice.” Psychological Review, 100(3).
Dreyfus, H. & Dreyfus, S. 1986. Mind over Machine. Free Press.
Shiner, L. 2012. “‘Blurred Boundaries’?” Philosophy Compass, 7(4).
Martin, R. 2009. The Design of Business. Harvard Business Press.
Erickson, K. 2024. “AI and Work in the Creative Industries.” Creative Industries Journal.
Hernandez-Ramirez, R. & Batalheiro Ferreira, J. 2024. “The Future End of Design Work.” She Ji, 10(4).
Eser, A. & Altiparmakogullari, Y. 2026. “Changing Skills of Industrial Designers in the Age of GenAI.” DRS 2026.