The observation
Two days ago, I published a piece on the two-image rule. The piece was about the mechanics of the rule — the four failure modes I keep watching, the four moves I have come to hold instead. The piece I want to write now is about the why — the deeper layer, the part that makes the rule feel reasonable to the manager who sets it, and the part that makes the rule feel inevitable to the person who has to live under it.
I have been working on this layer for the better part of a week, in the cracks between the meetings. The rule makes three mistakes at once. They are not the same mistake. They are mistakes at three different levels of the work — at the level of the individual, at the level of the process, and at the level of the relationship between the manager and the team. None of them is visible from inside the rule. All of them are why the rule keeps failing in the way it fails.
Let me name them.
What the rule misunderstands, in three moves
### Overjustification, at the level of the individual
The first thing the rule misunderstands is motivation. The rule assumes that the worker's effort is a function of the carrot and the stick — that the harder the carrot, the harder the worker works. The model is, in the language of psychology, an extrinsic model of motivation: do this, get this, do more of this, get more of this.
The problem is that the model is wrong, and the wrongness is not a matter of degree. It is a matter of kind. Edward Deci and Richard Ryan, in the Self-Determination Theory they have been building since the 1970s, have shown, in study after study, that extrinsic rewards do not merely add to intrinsic motivation. They crowd it out. A person who used to enjoy the work for its own sake — for the love of the craft, for the pride of the ship, for the curiosity of the problem — will, in the presence of a contingent reward, gradually shift the source of the effort to the reward. The shift is invisible. The shift is, in the worst case, irreversible. The phenomenon has a name, in the form I keep coming back to: the overjustification effect. When an external reward is introduced for an activity that was previously intrinsically motivated, the activity comes to be seen, by the person doing it, as something done for the reward. The intrinsic motivation is, in the research, the part that disappears.
I have been watching this happen, in the team I work with, in slow motion. The designer who used to spend a Sunday afternoon fiddling with a campaign layout, because the campaign was interesting, has been told, in the new system, that the Sunday afternoon is the time to fill the buffer for next week's quota. The fiddling has not gone away. The fiddling has been re-categorized — from "the part of the work I love" to "the part of the work I do because I have to." The re-categorization is the overjustification effect, in the small. The designer is, by every visible measure, doing more of the work. The designer is, by the measure that matters most to me, doing less of the work that the work is for.
In the team I run, the only sustainable source of effort I have found is the intrinsic one. I have a colleague who came to this job because of a long-standing fascination with overseas consumer culture. I have another who came because they wanted to learn how a brand is built from the inside. I have a third who came because they had spent two years making content that no one outside their apartment saw, and they wanted to make content that the world saw. None of these people came for the compensation. None of them is going to be retained by the compensation. The compensation is the price of admission. The fascination, the learning, the wanting to be seen — these are the reasons the work gets done at the level the work needs to be done.
The two-image rule treats the price of admission as the substance. The rule assumes that the worker's effort is, in the worker's own reading of the effort, the answer to the question "what will I get paid for this." The worker's effort is, in the worker's own reading, the answer to a different question — "what is interesting about this problem." The rule asks the wrong question. The rule rewards the wrong answer.
### Redundancy, at the level of the process
The second thing the rule misunderstands is how the work actually works. The rule is built on the industrial assumption that time is the scarce input, and that the manager's job is to fill the time with the most efficient possible use of the time. The model is the assembly line. The model is the one Frederick Taylor drew, in 1911, in the steel mill.
The model is, for creative work, the wrong model. The model is wrong not because creative work is undisciplined, and not because creative people should be free to do whatever they want. The model is wrong because the unit of progress in creative work is not the hour, and not the artifact, and not the deliverable. The unit of progress is the experiment — and the experiment, in creative work, fails more often than it succeeds.
The viral post is, in the version of the team I work with, the visible tip of a long tail of invisible experiments. The 100 drafts that did not land. The 30 hooks that did not catch. The 10 angles that did not convert. Each of these is, by the manager's standard, a waste. Each of these is, by the creative's standard, the substance of the work. The work is not the 1 that lands. The work is the 99 that did not. The 99 are the conditions under which the 1 becomes possible. The 99 are the redundancy that the next 1 depends on.
The 99 failed experiments are, in the language I keep coming back to, the redundancy that the 1 viral post cannot exist without. The rule that eliminates the redundancy has, in the same move, eliminated the conditions under which the 1 can be made.
The two-image rule is the rule of the 1. The rule assumes that the work happens in the gap between "no image" and "two images," and that the gap should be filled with the minimum possible work. The rule treats every hour not spent on the deliverable as an hour of waste. The rule, in the language of the 16:00 break piece, is the rule that has decided the form is the work. The rule, in the language of this piece, is the rule that has decided the deliverable is the work. The deliverable is, in creative work, the byproduct. The experiment is the work.
When the rule fills the calendar with the deliverable, the experiment has nowhere to go. The experiment gets pushed into the cracks, and the cracks, when the rule gets tighter, are the first things to go. The team that used to spend a Friday afternoon reading the comment section of an overseas brand, looking for a tone, looking for a pattern, looking for a hook — that team is now spending the Friday afternoon shipping the second of the two images. The comment section is unread. The pattern is unseen. The next hook, when it would have arrived, does not arrive. The team, by every visible measure, has shipped the two. The team, by the measure that matters most, has shipped less.
### Trust, at the level of the relationship
The third thing the rule misunderstands is the relationship between the manager and the team. The rule assumes that the worker's effort is, in the worker's own reporting of the effort, an input that the manager cannot verify directly. The rule assumes that the manager cannot see the work, and that the work must therefore be replaced by a count of the work. The proxy, in the manager's reading, is the workaround for the absence of observation.
The proxy is, in the manager's reading, also a workaround for the absence of trust. The manager who cannot observe the work has, in the moment the manager sets the quota, made a decision about the worker — the decision that the worker's report of the worker's effort cannot be trusted without a count. The decision is, in many cases, the right decision to make. The worker may, in many cases, be the kind of worker who will underperform if the worker is not counted. The quota is, in the right case, the right tool.
The quota is not the right tool for the kind of team I work with. The team I work with is small — two people, in the smallest case; four, in the larger case. The team is small enough that the manager can see the work, in the literal sense — can sit in the room, can read the drafts, can ask the question, can watch the worker think. The team is small enough that the count is, in the manager's reading, redundant. The count is also, in the manager's reading, a signal — the signal that the manager has decided, in this case, not to trust the worker without a count. The signal is, in the worker's reading, the answer.
The signal is, in the worker's reading, expensive. The worker who used to bring a problem to the manager — "I am stuck on this layout, I do not know which of these three angles is the right one" — has, in the presence of the quota, learned to bring only the answer — "I shipped the two, here they are." The conversation about the angles has not happened. The conversation, in the worker's reading, is the part of the work that is not in the count, and the worker, who has to hit the count, has put the conversation in the crack. The crack, in the small team, is the place where the manager was supposed to be. The manager is not in the crack. The count is in the crack. The count is the manager, in the small team's reading of the manager.
The cheapest tool to mask the absence of trust is a count. The most expensive tool to replace the absence of trust, in the small team, is the same count — because the count is, in the small team's reading of the count, the manager's statement that the team cannot be trusted without the count.
The small creative team, in the version I am trying to articulate, is the team where trust is the cheapest possible tool. The two-person team, in particular, is the team where the cost of a count is the highest possible cost — because the cost of the count is paid in the loss of the conversation, and the conversation, in a two-person team, is the entire workflow. The conversation is where the angle gets chosen. The conversation is where the hook gets made. The conversation is where the work, in the version the team is capable of, gets done. The count, in a two-person team, does not just fail to add the work. The count, in a two-person team, subtracts the work.
The principle, in one sentence
Creative work is intrinsic. Creative work is redundant. Creative work is relational. A manager who has decided to count the artifacts has, in the same move, decided to give up on the intrinsic, the redundant, and the relational — and to settle for the form, in the hope that the form is enough.
I have been holding this principle for a while. The hard part is not the principle. The hard part is the meeting where the manager has to say: the work is intrinsic, the work is redundant, the work is relational, the count has not measured the work, and the count, in fact, has not let the work be done.
What I have come to do instead
The moves I have come to hold are the moves that protect the three layers I have just named.
Protect the intrinsic. I do not, as a rule, tie compensation to a quota. I do, as a rule, tie compensation to a goal — and the goal is set by the team and the manager together, at the start of the cycle, in a conversation that is the most important conversation of the cycle. The conversation is where the worker tells the manager what the worker wants to learn. The conversation is where the manager tells the worker what the work needs. The conversation is the workflow. The number, when it appears, is a number of the goal. The number is not a number of the deliverable.
Protect the redundancy. I budget, in each cycle, a number of hours that are explicitly not tied to a deliverable. The hours are for reading the comment section of an overseas brand. The hours are for trying a hook that the manager has not approved. The hours are for the 99 experiments that did not land. The hours are, in the manager's reading, the redundancy. The hours are, in the creative's reading, the substance. The hours, in the version of the system I am trying to hold, are a line in the budget, defended in the review, named in the conversation. The hours are the part of the system that says: the work is not the deliverable, the experiment is the work.
Protect the relationship. I have, in the small team I run, decided to trust the worker's report of the worker's effort. The decision is, in the small team's reading, the manager's most expensive gift. The decision is, in the manager's reading, the manager's cheapest move — because the alternative is the count, and the count, in the small team, is the most expensive move. The decision is the conversation. The conversation is, in the small team, the workflow. The conversation is the part the count cannot replace.
What I am not saying
I am not saying metrics are wrong, and I have said this in the pieces before. I am not saying that the intrinsic is the only source of effort — compensation matters, and the worker who is not paid is not the worker who will stay. I am not saying that the redundancy is always good — sometimes the 99 experiments are a sign of a worker who is afraid to commit, and the conversation, in that case, is the conversation about the fear, and the conversation is the work.
I am not saying the manager has no role. The manager has the most important role. The manager is the person who sets the goal. The manager is the person who defends the room. The manager is the person who, in the meeting, has to say: the form is not the work, the room is the work, the hours in the chair are not the production, the production is what the team does with the room. The manager, in the version of the system I am trying to hold, is the most important person in the room. The manager is also, in the same room, the person who is most tempted to reach for the count.
I am not saying this is original. The argument is older than I am. The hard part is the discipline to apply the argument in the meeting, in the form, in the language the system speaks — and the discipline to keep applying it after the meeting is over.
Where this connects
The two-image-week piece is the direct predecessor. That piece was the mechanics. This piece is the why. The two pieces are the same argument at two different depths — the mechanics of the rule, and the reason the rule keeps being set.
The KPI piece is the structural case for the manager. The piece argued that KPI is a workaround for the manager who cannot see the work, and that the workarounds tend to fail in two predictable ways. This piece is the same argument, applied to a deeper layer — the layer where the workaround fails not because the count is the wrong count, but because the count is the wrong kind of thing to count.
The dirty-work piece is the closest cousin. The argument there was that dirty work belongs to the system, not the person. The argument here is the same argument, applied to the kind of work that the count destroys — the work that is intrinsic, the work that is redundant, the work that is relational. The dirty work, in this piece, is the count. The count is the dirty work. The count is the part the system should be doing, in the background, in the form of a goal and a conversation. The count, in the version of the system I am trying to hold, is not on the creative's plate.
The 16:00 break piece is the daily version of the redundancy argument. The 16:00 break is the half hour of redundancy the manager has decided to protect. The 16:00 break is, in this piece's reading, the smallest possible version of the buffer. The buffer for the 99 experiments is the larger version. The two are the same argument at two different scales.
What the literature already says
I did not invent any of this. The framing is borrowed, openly, from work that has been around much longer than I have.
Edward Deci and Richard Ryan, Self-Determination Theory. Deci and Ryan's research, in the form I keep coming back to, distinguishes between intrinsic motivation (the motivation that comes from the activity itself) and extrinsic motivation (the motivation that comes from an external reward or punishment). Their work, in the four decades since it was first published, has shown, in study after study, that extrinsic rewards crowd out intrinsic motivation when the rewards are introduced for activities that were previously intrinsically motivated. The phenomenon is, in the language I quoted earlier, the overjustification effect. The implication, for the manager of a small creative team, is the punchline of this entire piece.
Daniel Pink, Drive. Pink's argument, in the popular form I keep coming back to, is that the work of the 21st century is built on three intrinsic motivators — autonomy, mastery, purpose. The two-image rule, in Pink's reading, attacks all three. The rule removes the autonomy. The rule substitutes mastery with a count. The rule, in the worst case, makes the purpose the count itself. The worker, in the rule's reading, has no autonomy, no mastery, no purpose — only the number. The worker, in Pink's reading, is the worker who has been hollowed out.
Teresa Amabile, The Progress Principle. Amabile's research, in the form I keep coming back to, is that the single biggest motivator of work, across thousands of journal entries from real workers, is the sense of making progress — the small, daily, often invisible sense that the work is moving. The two-image rule, in Amabile's reading, has a curious effect: it creates a visible sense of progress (the two, shipped) and a felt sense of anti-progress (the intrinsic motivation, the redundancy, the relationship, all eroded). The visible progress is, in Amabile's reading, not the same as the felt progress. The visible progress is the count. The felt progress is the work. The count, in the worst case, has the visible progress and not the felt progress. The count, in the worst case, has the form and not the work.
I am not going to re-explain them here. The point is the same point.
A postscript
The hardest part of being a manager, in the version of the job I am trying to hold, is not the work. The work is mostly learnable. The hardest part is the moment the manager is asked, by the system, to defend the no-count, the no-quota, the no-form — and the manager has to defend, in the language the system understands, the decision to trust the team. The system, in my experience, often does not understand. The system, in my experience, often wants the form. The manager, in the version of the job I am trying to hold, has to say no. The manager has to say no, in the meeting, in the form, in the language the system speaks.
I am not always sure I say no well. I am getting better at it. The improvement is the only thing I have to point to, at the end of the year, as the work I have actually done.
I am going to keep working it out, in the systems I build, in the meetings I have, in the small decisions about whether to count or to trust. I am going to keep working it out because the alternative is the form, and the form is not the work.
The hardest part is not the principle. The hardest part is the moment the system asks, and the manager has to say no, and the manager has to be ready to defend the no, in the language the system understands, in the meeting the system has scheduled, in the form the system has decided is the form.