Most of you will probably know that Paul Rand is the designer of the IBM logo, the ABC logo, the UPS logo, the Enron logo, the NeXT logo, the Westinghouse logo, etc. A titan! And maybe an organizational theorist? His 1987 essay "Good Design is Good Will" is, underneath its aesthetic philosophy, an argument about structure.
The essay is about the gap between designers and management. Rand identifies three causes of poor design:
- Management's indifference to quality;
- Market researchers' vested interests; and
- Designers' lack of authority or competence.
Buried in the middle is this sentence:
"I believe that design quality is proportionately related to the distance that exists between the designer and the management at the top. The closer this relationship, the more likely chances are for a meaningful design."
OD!
What should we take from this?
- Proximity to decision-makers determines quality. The closer the practitioner is to leadership, the less the work is filtered through layers of opinion from people without relevant expertise. Rand illustrates this with Hermann Bahlsen and the Bahlsen biscuit company—a manufacturer who, in Rand's telling, "combined art and his work in the most thorough fashion," sustaining a direct relationship with designer Martel Schwichtenberg over time and producing work of consistent excellence.
- Committee work degrades creative output. "Group design or design by committee, although occasionally useful, deprives the designer of the distinct pleasure of personal accomplishment and self-realization" and prevents ideas from having "time to develop." This is about the kinds of conditions that produce good work.
- Management unawareness is structural. People who approve design work usually haven't been trained to assess it. Rand is careful not to be contemptuous here; he notes that "lay people who have an instinctive sense for design... leave design to the experts."
- Design is an ethical practice. "A badly designed product that works is no less unethical than a beautiful product that doesn't." Whether organizations can develop something like taste (felt judgment even over criteria) is a related question.
The org design angle
Rand's core argument maps onto Expanded Available Power, or the idea that organizations generate better outcomes when they create more power to distribute throughout the organization, rather than concentrating all of it at the top. The designer-to-management distance Rand describes is a proxy for exactly this. When design decisions travel up a long chain of people without design backgrounds, power from expertise into preference, from quality into safety.
Rand's own practice is instructive here in ways he didn't quite acknowledge. His IBM work was a group project, and it predates the essay by thirty years. Eliot Noyes assembled the program starting in 1956, with Rand for graphics, Eero Saarinen for architecture, Charles and Ray Eames for exhibitions and films. Each operated with real authority in their domain, or a working example of Emergent Leadership, where authority follows expertise even over title. They designed in parallel, under shared principles, with a client relationship close enough to keep everything cohesive. This was what we'd now call a federated structure, and by the time he wrote the essay, it had three decades of results behind it.
The practice Rand was observing in 1987—mostly focused on graphic design, logos, print—has since expanded into something he probably wouldn't have anticipated. The design of a major digital or physical product today (happily!) involves researchers, interaction designers, visual designers, content strategists, accessibility specialists, and engineers, all making decisions that compound over time across millions of touchpoints. The scale is categorically different, and the shape of the federation (if not the size!) needs to match.
The design systems movement is one answer. The model, developed explicitly at companies like Google and Salesforce, and now the de facto standard for any design org of meaningful size, works like this: a small, usually seniorish core team sets direction and maintains standards; contributors embedded in product teams participate in evolving the system without ceding their autonomy. It's neither individual authorship nor design-by-committee. Rand would recognize the instinct to preserve the integrity of judgment and resist dilution, even if the structure is new. He just had a logo where we have a living codebase.
What the design systems movement has built is something close to the Guilds pattern: knowledge networked across a distributed organization, maintained through shared standards rather than shared location or reporting lines.
Rand's core intuition is unchanged: distance between designer and decision-maker is where quality goes to die. In a federated design organization, the relevant distance is between the standards-setting function and the people executing against it. Keep that distance short and the relationship substantive, and scaled coherence is the result. Let it grow, and you get what Rand feared: aesthetic judgment replaced by political judgment, quality replaced by safety.
"Any system that sees aesthetics as irrelevant, that separates the artist from [their] product, that fragments the work of the individual, or creates by committee, or makes mincemeat of the creative process will, in the long run, diminish not only the product but the maker as well."
Source: Paul Rand, A Designer's Art, 1985