Taste Cannot Be Delegated
You can outsource almost everything. Not the knowing.
By Mark Andrew Palmer · · 8 min read
Somewhere in the last decade, taste got rebranded as a bottleneck. The industry decided the fix was to distribute the deciding. You cannot distribute the deciding.
Somewhere in the last decade, taste got rebranded as a bottleneck.
The complaint goes like this. The work is piling up. The creative director insists on seeing everything before it ships. The reviews take too long. The founder keeps sending things back with notes nobody else would have written. The organization, being an organization, diagnoses this as a process problem and prescribes the usual medicine. Distribute the deciding. Empower the team. Build a brand guidelines document thick enough to stop a door and let the system do the judging.
And then, reliably, quietly, over about eighteen months, the work goes gray.
Nothing is wrong with it, exactly. Everything clears the checklist. The logo is the right size. The tone of voice matches the tone of voice document. Every individual piece is defensible, and the total is forgettable, and nobody inside the building can figure out why, because every box got checked by a competent person acting in good faith.
Here is why. Taste cannot be delegated. It was never a process. It was always a person.
A guidelines document is a photograph of taste. Useful, flattering even, and completely unable to make the next decision.
Mark Palmer
Let me define the term, because the word gets used loosely and I intend to use it precisely. Taste is not preference. Preference is liking blue. Taste is knowing, in front of two options that both technically work, which one is alive and which one is merely correct, and knowing it fast, and being right at a rate that cannot be explained by luck. It is pattern recognition compounded over decades, stored somewhere south of language, retrieved in about half a second, and almost impossible to justify in a meeting, which is exactly why meetings kill it.
Taste is the residue of ten thousand judgments that got tested against reality. The gallery shows that taught you what dies on a wall. The campaigns that taught you what dies in the market. The typefaces you loved at thirty and cannot look at now. Every one of those deposits lives in one nervous system. There is no export function.
The delegation fantasy fails for a simple structural reason. Taste is not a set of conclusions. It is the faculty that produces conclusions. When you write the guidelines document, you are recording the conclusions taste has already reached, past tense. The next decision, the one that arrives Tuesday and resembles nothing in the document, still requires the faculty. The document will sit there, thorough and useless, while a committee votes on which of three adequate options feels safest. The committee will choose the safest one. Committees always do. That is not a character flaw. It is arithmetic. Consensus is an averaging function, and taste is the refusal to average.
This is the actual reason the creative director must stay in the work, which I have argued elsewhere on this site and will keep arguing until the industry stops needing to hear it. Not ego. Not control. Physics. The judgment lives in a person, so the person has to be in the room where the judgments happen. Every layer you insert between the taste and the work is a layer of averaging, and enough layers will average anything down to beige.
The AI moment has made this arithmetic brutally visible, and I say this as someone who uses the tools daily and without apology. Generation is now effectively free. Anyone can produce a hundred options before lunch. Which means the entire remaining value of a creative professional is concentrated in the act the machine cannot perform: looking at the hundred and knowing which one, if any, deserves to exist. The scarce thing was never the making. It was always the knowing. The tools did not threaten taste. They repriced it. Sharply upward.
When everyone can make everything, the only question left is who knows what is worth making. That question has never had a committee for an answer.
Mark Palmer
Abraham Maslow got to this conclusion before any of us, and he got there with data. In the revised edition of Motivation and Personality, published in 1970, the year he died, he wrote that “how a man thinks can’t be separated from what he is and the question of what he thinks he is is never independent of what he is, in fact.” The judgment and the person are one instrument. And then he went further, into territory the branding industry has still not caught up with. He argued that the most developed people among us can be used, his word, as assayers.
It has been my experience, through a long line of exploratory investigations going back to the 30’s, that the healthiest people, or the most creative, or the strongest, or the wisest, or saintliest, can be used as biological assays, or perhaps I should say advanced scouts or more sensitive perceivers to tell us less sensitive ones what it is that we value. What I mean is something like this: it is easy enough to select out, for instance, persons who are aesthetically sensitive to colors and forms, and then learn to submit ourselves or to defer to their judgements, about colors, forms, fabrics, furniture and the like. My experience is that if I get out of the way and do not intrude upon the superior perceivers, I can confidently predict that what they like immediately, I will slowly get to like in perhaps a month or two. It is as if they were I, only more sensitized, or as if they were I, with less doubt, confusion and uncertainty. I can use them, so to speak, as my experts, just as art collectors will hire art experts to help them with their buying. This belief is supported by the work of Child, which shows that experienced and expert artists have similar tastes, even cross-culturally. I hypothesize also that such sensitives are less susceptible to fads and fashions than average people are. Now in this same way I have found that if I select psychologically healthy humans, what they like is what human beings will come to like. Aristotle is pertinent here: ‘what the superior man thinks is good, that is what is really good.’ For instance is empirically characteristic of self-actualizing people that they have far less doubt about right and wrong than average people do. They do not get confused just because 95% of the population disagrees with them. And I may mention that at least in the group I studied they tended to agree about what was right and wrong as if they were perceiving something real and extra-human rather than comparing tastes that might be relative to the individual person. In a word, I have used them as ‘value-assayers’. Or perhaps I should better say that I have learned from them what ultimate values probably are. Or, to say it in another way, I have learned that what great human beings value are what I will eventually agree with, what I will come to value, and what I will come to see as worthy of as valuable in some extra personal sense and what data will eventually support.
Abraham Maslow, Motivation and Personality (1970)
Read that again and notice what a research psychologist is admitting. The superior perceiver is not expressing a preference. They are detecting something. What they like immediately, the rest of us come to like in a month or two, and the agreement runs cross-culturally, as if taste were perceiving something real rather than voting on something relative. Maslow did not survey a committee to find out what humans value. He found the most sensitive instruments available and deferred to their readings. He hired taste. He did not distribute it. The whole delegation fantasy dies right there, in a footnote from 1970, and the industry has spent fifty years pretending it never read it.
Now, the part that sounds like a contradiction and is not. Taste cannot be delegated, but it can be developed. Those are different operations entirely. Delegation tries to move the faculty from one head to another by memo. Development builds a new instance of it, slowly, in a new nervous system, the only way one has ever been built: by making thousands of judgments next to someone whose judgment is better than yours, getting overruled, and paying attention to why.
That is apprenticeship, and the industry mostly dismantled it because it does not scale. The ateliers knew. The Japanese workshops still know. You do not hand a junior the guidelines document and wish them luck. You let them watch a decision get made, then another, then ten thousand more, until one day the overruling stops because it has become unnecessary. What transferred was not the conclusions. It was the calibration. That takes years, it resists every efficiency the org chart can throw at it, and it is the only method with a working track record in the entire history of craft.
So here is the uncomfortable position I am asking you to consider, whether you run a company or a studio or a body of work that currently consists of you and a glass of something good after midnight.
Somebody’s taste is going to govern the work. That is not optional. The only question is whose, and whether it is any good. Refuse to name the person and you have not escaped the tyranny of taste. You have handed the job to the averaging function, and the averaging function has terrible taste, uniformly, everywhere it has ever been given the chair.
At Omythic the answer is not ambiguous. You work with the person whose name is on the judgment. The bottleneck is the product. I have watched what happens when the deciding gets distributed, because I built a sixty-eight-person machine that distributed it beautifully, and I have written the autopsy elsewhere. The work went gray at exactly the rate the deciding spread out. That is not a coincidence I intend to repeat.
Keep your taste in the work. Grow more of it the slow way, next to you, in people who will one day overrule their own juniors for reasons they also cannot fully articulate.
But do not delegate it. It does not delegate.
It dies in the handoff, quietly, and the checklist never notices.
Tags: Creative Process, Branding, Taste, Creative Direction, Leadership, Craft, Decision Making, Omythic, Mark Palmer