Antifragility > understanding

Robot avatar images lovingly delivered by Robohash.org.

Winkletter  •  22 Nov 2022   •  

I’ve been attempting to understand an article written by Nassim Taleb: Understanding Is a Poor Substitute for Convexity (Antifragility). I think I finally have a personal example that makes sense to me, and answers a question I have.

Why isn’t the coloring book market flooded with the work of professional artists?

Some artists do publish coloring books and make successful careers in the niche. But there are hundreds of thousands of artists who are capable of producing coloring books and they don’t. Instead, many of the coloring books on sale at Amazon are created by people with little or no understanding of art. What’s up with that?

The reason why artists don’t make many of the coloring books is because the market is highly variable, which means that convexity beats understanding.

Let’s say you’re a professional artist and you want to make a coloring book with 50 illustrations. It takes you 8 hours to make one illustration, and 400 hours to make one coloring book. Because you’ve invested two months of work in this book, you need to charge a premium. At the price you’re asking, customers expect high-quality paper and printing. This requires a print run and an up-front investment of capital, plus storage costs to house the physical copies at the Amazon warehouse.

This model does not thrive on variability. So, you end up paying Amazon more money to advertise your coloring book and bring down some of the variability. That variability being “who sees your book.” But there’s still a lot of variability around “what customers want.”

Meanwhile, I’m a non-artist. I may hire a group of freelance artists, or use clip art libraries, or purchase interiors to make my coloring books. Let’s say I can make 20 coloring books in the time a professional artist makes their one. I upload my files to KDP.

  • 12 of my books get little or no views and no sales.
  • 5 of my books get a handful of sales and reviews.
  • 2 of my books do well.
  • 1 book does exceptionally well.

I make bank because my process thrives on variability. If there were no variability because every book was seen by a requisite number of people and purchasing decisions were made on quality alone, then professional artists would flood the market. And all coloring books would cost $15.

In this environment, understanding will be beaten by convex trial and error. The convexity in this case refers to the number of books I produce and the number of sales that result. The more I produce, the more likely one of those books will resonate with consumers. At a low number of book sales are low. At a higher number, the sales are much much higher. This creates a convex curve.

With each book I make I tinker a bit with the process.

An artist understands art. They study coloring books and practice making beautiful art that is a pleasure to color. They learn how to create a coloring book that is a treasure to behold. Nobody buys it. Why? Consumers are fickle. No matter how much you understand, you won’t understand what they want until they start throwing money at you.

That being said, I’ve finished the cover for my next coloring book. We’ll see if this one resonates with people.

dapper dogs cover

Comments

Love this exploration! Always been a fan of Nassim Taleb’s work, even though I often struggle to understand his prose. I think the variability and convexity you mentioned are exactly why I run a portfolio of projects! I don’t always understand how or why 1 product works, but with enough attempts on goal, 1 or 2 will appear.

jasonleow  •  22 Nov 2022, 11:35 pm

Hopefully I’m not misrepresenting his work.

It seems like your Caard plugins fall under this category of antifragile/convexity. There’s variability and uncertainty around what plugins people want. You can’t know what projects people are building and what functions they need so you make a lot of small bets. If there were no variability, Caard would just add the functions people need to the core application. The variability is probably a bit frustrating when you’re trying to understand what people want, but that variability is the main feature that makes that project worthwhile.

Robot avatar images lovingly delivered by Robohash.org.
Winkletter  •  23 Nov 2022, 4:23 am

@Winkletter that’s a great assessment of why the plugins work 🔥

jasonleow  •  24 Nov 2022, 2:43 am