Iterative cover design: how my friends and readers guide the visuals
I have a confession to make: I design my own book covers, and they are, from reader feedback, pretty good. Nothing that will win design awards, but functional stuff that works, like this:
But I’m not a designer.
I cut my teeth as a writer, as a blogger and tech journalist, and now as a data-scientist-in-training. What I’ve come to realize, by hanging out and getting plastered with a lot of folks from the ad industry, is that drawing, design and art is a different ball game. Very few people know how to make a bad cover good, or to suggest changes that can take a piece of design from passable to good. People either have this design sense . . . or don’t.
When I was working on Numbercaste, I worked with a professional artist. Something must have gotten lost in the communication, because over multiple conversations we produced this:
Now, I thought this was epic. I really liked it. But when I showed it to my friends, they weren’t impressed. I initially wrote this off as being subjective. We know design is a highly subjective thing. One man’s trash is the next man’s treasure, so to speak, and I just thought this wasn’t right for them.
But the more people I showed it to, the more I realized that the votes tallied up in a very bad way. Yudha + artist = yay. Thirty other people = nay.
At this point I conceded that they had a point. There is a thing called the Sunk Cost Fallacy: the more time and effort we spend on something, the more invested we are in it. And this means we often miss better choices or continue past the point of diminishing returns. I’d sunk three months of time into that first cover, and I was seeing the Fallacy in action.
So as an experiment I started posting covers online. Stuff I had created. What I realized was simple: individually, we may not be star designers, but if you take the most common gripes and feedback and lay them out, you have an understanding of what’s right or wrong. One person might be biased or wrong. Thirty, a hundred people, when compared with just my opinion, are highly likely to be right.
And once I had decided that I would let my friends judge this process, it led me from disasters like this (which I thought was very cool):
To what you saw at the start: readable, engaging, functional.
My metrics were to look at the attention each iteration received – even the number of people liking one over the other mattered: and now, with Numbercaste doing pretty well for a debut, one of the most consistent pieces of feedback I get is: “Nice cover!”
I tried again with my new book.
These were covers I thought were greatest (the ones I had spent the most amount of time on):
But what my friends liked was something I had spent very little time on, but was striking the right chord. Reading all of their feedback, this we iterated into:
Which is gorgeous, and would never have occurred to me at the start, because I was just mucking around with some other stuff. People liked the split-mandala. People also liked the gold and the black. Simply by tallying votes for different options and comments I could figure out what it was that worked best in each design. A lot of indie authors do this: we have groups where we post covers and ask for feedback.
My takeaway is simple. Most of the people I interact with are very well-read. They know a good book cover when they see one. And they are also my first readers and the people who reach out about my work the most. If I can trust them to read something I write, I can also trust their opinions when it comes to making that book better. Especially if it’s something where I have little innate sense of how things should be, like design.
People have actively changed these covers that you just saw: Thorn Coyle, Vera Brook, Jessica Ferrari, Pamudu Tennakoon, Govind Dhar, Sofie Couwenberg, Yusra Aziz-Eliyas, Louiqa Raschid, Shabeeb Muzammil, Vishva Herath, Nigel White, Anudi Nanayakkara, Dilina Amaruwan, Aisha Nazim, Nisansa de Silva, Tristan Vick and Jessica Perera. When I look at what they have to say on these matters, it often turns out far more accurate than my own judgement or that of any one person I consult.
So try and replicate this: ask for feedback. You don’t have to do your own covers; that’s simply my story and my way of working. It might be that you have professional covers lined up. Nevertheless: ask for feedback from the people who will read your work. Look at all the responses. Aggregate. Iterate. It works for me, and it might work for you as well.