Two Essential Studies for Makers
On consistency, novelty, and variety
In the writing of this blog, there are two academic papers I find myself returning to again and again. Together, they offer a unified theory of what predicts creative success that is backed by more than self-report. At risk of overgeneralization: I believe these studies hold the key to systematic understanding of who wins and loses in domains as diverse as art, academia, and entrepreneurship.
The first paper, written by psychologist Bob Boice, is called “Which Is More Productive, Writing in Binge Patterns of Creative Illness or in Moderation?” In this study, subjects (junior faculty members in the humanities, social sciences, and hard sciences) were divided into two groups: “binge writers” and “regular writers.” The bingers were those who tended to write in great bursts of output followed by long periods of inactivity. (They also tended to report feelings of euphoria, were less likely to rely on notes/outlines, and displayed a greater frequency of errors as evidenced by deletions or erasures mid-writing.) The regulars displayed the opposite pattern: brief daily sessions of relatively planned, sedate output.1
At the end of two years, Boice looked at the results of these two groups of writers. The findings were unequivocal: Regulars had more success than bingers. This was obvious both in the volume of work they produced and the number of manuscripts they had accepted for publication.2 Additionally, outside of their euphoric writing sessions, bingers displayed many more symptoms of depression, as measured by the Beck Depression Inventory.
There are some reasons to take Boice’s conclusions with a grain of salt. First, as his title suggests, he is clearly coming at the issue from a strong perspective. (The very term “binge writing” is a slight against more burst-y modes of output.) At just sixteen subjects, the sample size of the study is quite small, and participants were not randomized but identified in advance as one of the two types. Finally, it’s worth emphasizing that the writers in question were academics—not, e.g., abstract painters or experimental poets—which might lead some to dismiss the generalizability of the results to other creative pursuits.
Still, even if the study does not demonstrate true causality across all domains, it is highly suggestive. In particular, it should give us pause before we blindly accept the most popular cultural scripts about creativity—“the romantic view [that] creativity depends on excess and disorder” and that “a well-organized and sane mode of working at writing would undermine brilliance.”
The romantic view is dangerous because it suggests that the things which make us sick or undermine our happiness are necessary for genius. And perhaps it’s even true that a bit of struggle and unpredictability are necessary ingredients for innovative work! But personally, I’ve never found our world to be lacking in chaos or difficulty.3 It’s for this reason that I find Boice’s ultimate conclusion highly persuasive:
[C]reatives often find success in spite of the excessive emotions and distractions they generate. But they could…work even more productively and creatively in the long run with the efficiencies of moderation. [emphasis mine]
The second paper, by organizational behavior and management researcher Justin Berg, is One-Hit Wonders versus Hit Makers: Sustaining Success in Creative Industries. In contrast to the small sample size of Boice’s study, Berg boasts an impressive volume of data: He tracks 3 million songs by 69,050 musical artists, from over five decades, to draw empirical conclusions about those artists’ careers.
From this huge volume of data a few things become clear. First, as with manuscript acceptance in Boice’s study, the raw rates of success are remarkably low. In a domain as competitive as music, it’s hardly surprising that artists struggle to distinguish themselves from the crowd—but the actual numbers are striking: Twenty years into a music career, the odds of any artist having a top-100 hit max out at about twelve percent. (And that just includes people in Berg’s sample—i.e., those who actually signed with a record label.)
However, like Boice, Berg also identifies different modes of output that vastly change the likelihood of success in percentage terms. Specifically, he looks at artists who are high or low in novelty—i.e., how different their songs are from other people’s; and high or low in variety—i.e., how different their songs are from each other.4
What he finds is that at any stage in an artist’s career, greater variety is good, but greater novelty presents a tradeoff depending on whether the artist already has a hit.
Before an artist is discovered, variety is a boon because it gives them more bites at the proverbial apple—a broader portfolio of creative “investments” which might get hot. Novelty, on the other hand, serves as a disadvantage, which makes sense: By definition, the artist is doing something weird, so they’re less likely to find success than their peers producing more conventional songs.
After an artist has achieved initial success, variety remains a positive predictor, since it gives the artist greater optionality in adapting to trends or changing tastes. With novelty, the story is reversed. Conditional on initial success, it is good to be unconventional. This also makes sense: If you’re doing something weird that people happen to like, you have effectively carved out a niche where you are the dominant or only player.
Crucial to Berg’s model is the idea that the moment of breakout success is effectively a crystallizing moment—one which locks in whatever songs are currently in the artist’s portfolio. (He argues that this crystallization is due to both external expectations and internal learning along the lines of “This is what’s worked for me so far.”) Whatever the reason, artists will be punished in the marketplace for producing songs that are not consistent with their existing work.
Strategically, these findings counsel one clear strategy and one tough choice: Early on, creatives should try to produce a variety of work, which both increases the odds of initial success and helps to sustain success; but the decision to produce novel work is high-risk, high-reward—simultaneously increasing the odds of both sustained success and no success at all.
I find these two studies remarkably insightful and practical. However, I don’t think they should be taken to suggest that creativity is a math problem or reducible to a formula. Call it flow, call it the muse—there is undeniably something mystical in the creative act: the way life seems to emerge spontaneously from the primordial soup of the unconscious. As writer Lan Samantha Chang describes the experience:
Inside this walled-off room, time is different—it is flexible, malleable. We’re allowed to bend it, to speed it up, slow it down, to jump forwards and backwards, as our minds do. We can circle back to our thoughts and memories picking and choosing the most meaningful to us. There’s a hushed, glowing sound, like the sound coming from the inside of a shell.
Unfortunately, this sacred space of the imagination is not the world most of us inhabit most of our waking hours. In the external world—the world of glass and concrete, commerce and statistics—time is all-too-linear, and the sounds of shells are drowned by car horns and alarm clocks. In the external world, we are tracked and measured, cogs in the economic machine. We ignore the lessons of this world to our peril, for its walls contain the universe that swirls inside our heads.
More precisely: “Members of the binge group wrote with no more than one significant interruption or pause (but otherwise with a chronically fast pace) for at least 2 hours, usually far longer (range = 2 to 12 hours). Regular writers generally wrote for no more than 1 hour each day (and with a dominance of moderate pacing plus at least four pauses for reflection/relaxation), rarely for more than 90 minutes.”
In raw numbers, the publication rate for both groups is small (this is, after all, over just two years in academia); but as a percentage increase, the regulars outperformed the bingers by many multiples.
Flannery O’Connor: “Anyone who survived childhood has enough material to write for the rest of his life.”
Berg scores these qualities by looking at song features like “acousticness” and “danceability.”