I let ChatGPT pick my novel’s comp titles

Today I came across this article on Jane Friedman’s blog about what to type into Chat GPT to let artificial intelligence generate comp titles for you.

For those of you who don’t know (bless you), comp titles are recently-published books comparable to the one you’re trying to get published.

I spent God knows how many hours looking. I asked my writer friends what they thought would make good comps, and ended up using those and a few of my own (which, admittedly, weren’t really good).

If only I’d read about ChatGPT months ago!

Comp title? WTF? Just publish my book!

Even though I’m done with the querying process (hallelujah!), I had to check it out. So here’s what I asked Chat GPT:

“Suggest 10 comp titles published in the last three years for a literary campus novel set in Vermont that features philosophy. Do not include any books that have appeared on bestseller lists. Do not include nonfiction books.

And my results:

“Sure, here are 10 comp titles published in the last three years for a literary campus novel set in Vermont that features philosophy, excluding bestsellers and nonfiction books:

  • “The Philosopher’s Flight” by Tom Miller (2018)
  • “The Idiot” by Elif Batuman (2017)
  • “The Shakespeare Requirement” by Julie Schumacher (2018)
  • “The Ensemble” by Aja Gabel (2018)
  • “The Female Persuasion” by Meg Wolitzer (2018)
  • “The Overstory” by Richard Powers (2018)
  • “The Red Word” by Sarah Henstra (2018)
  • “The Dreamers” by Karen Thompson Walker (2019)
  • “The Incendiaries” by R.O. Kwon (2018)
  • “The Lonesome Bodybuilder: Stories” by Yukiko Motoya (2018)

I hope this helps!”

Of the above, the only ones I considered were The Overstory and The Idiot, but I decided against them. Overstory keeps popping up on my Amazon recommendations, so maybe ChatGPT is onto something.

I clicked “Regenerate” and I got these completely different results:

  1. “The Philosophy Club” by Kimbrielle Hall: A group of students in a small Vermont college form a philosophy club and their discussions lead them into deeper existential and ethical questions.
  2. “The Missing Course” by Julia E. Antoine: A philosophy professor disappears, leaving his students to navigate their thoughts and emotions about his disappearance while they continue to study philosophy on their own.
  3. “The Field of Questions” by Alyssa P. Feirman: A young philosophy professor at a liberal arts college in Vermont struggles with the weight of her own philosophical inquiries while also facing personal and professional challenges.
  4. “The Ripple Effect” by Miranda A. Davis: A campus novel set in a Vermont college that follows the relationships and philosophical discussions between a group of students and a beloved philosophy professor.
  5. “The Examined Life” by Caitlin M. Todd: An introspective campus novel set in Vermont that follows a young woman’s journey through her college years as she grapples with existential and philosophical questions.
  6. “The Art of the Argument” by Rachel N. Wong: A philosophy professor at a small Vermont college takes her students through the art of argumentation and reasoning, but finds herself in a moral dilemma when one of her students takes it too far.
  7. “The Ethics of Ambiguity” by Samantha R. Sanchez: A campus novel set in a Vermont college that follows a young woman as she navigates her way through philosophical questions and the complexities of relationships and social hierarchies.
  8. “The Philosophy of Love” by Jordan C. Rodriguez: A group of students at a Vermont college explore the philosophical concepts of love and desire, and how these concepts shape their relationships with each other.
  9. “The Language of Reality” by Stephanie L. Green: A philosophical campus novel set in Vermont that explores the idea of language as a tool for understanding and interpreting reality.
  10. “The Examined Mind” by Ethan R. Black: A young philosophy professor at a small Vermont college grapples with his own philosophical beliefs and tries to help his students do the same, while also navigating the personal challenges of academia.

Wow. Sounds really promising, right? Why haven’t I heard of any of these books? Here’s why:

  1. I looked up the first book, “The Philosophy Club”, first on Google, then on Amazon, but I couldn’t find it. I looked up “Kimbrielle Hall.” Nada. So much for that.
  2. Julia E. Antoine is an author who writes philosophy books, but I couldn’t find the title mentioned by the AI.
  3. Nada.
  4. Nothing.
  5. Nope.
  6. I’m starting to get the idea and I’m giving up.

That said, those invented books sound pretty damned interesting! Too bad they don’t exist…yet!

I think ChatGPT might have to write my next query letter for me. It’s doing a better job than some of my first attempts (and no, I won’t share).

In case you’re interested, here are my actual comp titles:

A FOOTNOTE TO PLATO is written in the tradition of Coetzee’s Disgrace, Roth’s A Human Stain, and David Lodge’s Campus Trilogy, with shades of The Chair and Joshua Cohen’s The Netanyahus. 

I know it seems ridiculous to get so worked up over comp titles, but if you’ve ever had to do it, you know what I mean. For those of you who don’t know, you are expected to spend months doing research on this so your editor can justify taking you on to the publisher. In other words, you have to get business minded in a way that’s not even remotely fun.

I started out with different titles, but eventually came to the ones above. But they aren’t great; they were what I settled on because I got so sick of the whole business. Here are their problems:

Coetzee’s Disgrace makes sense, sort of, but was it written in the past three years? No.

Human Stain? It works, and I even mention the book in my novel, but it’s not recent.

Campus Trilogy, ditto. Actually, I don’t see this one as much, but it’s okay. Still, same time problem.

The Netanyahus is where things get interesting. Two of my writer friends who don’t know each other both recommended this book as a comp title. I had read the book, but I never would’ve picked it. If I squint, I can see it. I see some stylistic similarities, particularly in Cohen’s comedic dialogue and timing. Our books ask readers to think for themselves when presented with opposing views. Our books are about professors. But Cohen’s book has no philosophy in it, and that, I would think, would disqualify it. Still, when two writers independently recommend it, I take heed.

The Chair isn’t a book (I don’t think?), but I think it does the trick.

Other suggestions from writers: Several books by Amor Towles. Two writers suggested him, but not independently. I felt awkward including any of his books, (they picked “Rules of Civility” and “Lincoln Highway”) even though I did study his opening to A Gentleman in Moscow to help me with transitioning in omniscient. Why did I feel strange about this/these picks? Well, stylistically, maybe they’re right. I loved Gentleman in Moscow, so why wouldn’t some of his style seep in? In particular, one writer noted that I tend to end my chapters on a thematic or resonant note, as does he. (I thought that was what you were supposed to do?) Well, I was flattered, but I had to admit that there weren’t many thematic similarities, and besides that, I feel like he’s too good to compare myself to. (BTW, I later read their recommended books, Civility and Highway…love love love.)

How about you? Have you had to come up with comp titles? Or have you tried using Chat GPT? What did you use it for? What were your experiences?

14 thoughts on “I let ChatGPT pick my novel’s comp titles

  1. I’ve used ChatGPT for a lot of things, from generating programs to writing stories, even to funny names or acronyms. I even used it to help edit a 50K novel I dusted off from a previous NanoWriMo. The experience was good, and while it wasn’t perfect, it was great and I was convinced it was a worthy tool. At one point, I was having it rewrite dialog to sound of a certain style, and it would remember character’s voices when I told it to plot out dialogues.

    Liked by 1 person

      • Well, I pasted sections from the work and asked it to do things like flesh out a particular scene or change the voices as mentioned below. Sometimes, I might ask it to tighten the wording, reword it in another author’s style or even add content for a scene.

        I even ask it to produce outlines or find gaps in a scene.

        More often than not, what it provides me is something I use to build upon. I rarely take what it writes verbatim, but it gives me enough to point me in what I should do, or even what won’t work, and that saves me a lot of effort.

        In my case, the editing was for something I did for fun, so I was content to use most of what it did verbatim, then later go back and clean it up. It actually turned out that with all the work I did in it, I pretty much ended up rewriting the novel largely with it’s help, even if I did do a fair bit of hand revision. It was far faster than if I did it manually.

        Liked by 1 person

  2. I tried reading the preface and first chapter of Kant’s “Religion Within the Boundaries of Mere Reason,” both of which totally eluded me, so I had ChatGPT try to summarize the key points. I got the main gist, but obviously the details—wherein is the Devil—didn’t come across.

    Liked by 1 person

    • Ha, nice! I can see the temptation to do that with Kant, but as you say, the devil being in the details would make me wonder if the AI was accurate. Maybe one day it will be reliable enough to get a Cliffs Notes version of difficult texts. Until then, I’ll just rely on you. 🙂


  3. Never had to come up with comp titles. Didn’t know that was a thing authors had to do. Yuck. (Although I did see something like it as a recommendation in one of the writer guides, as a form of market research.)

    I’ve played around with ChatGPT on an off over the last few months. But never thought to ask it for books similar to one I liked. Tried it after reading this post in terms of an author and series. The results weren’t bad. All the series existed, although it was about half and half (across two runs) on being what I would consider really similar. At least they were all in broadly the right genre (sci-fi).

    Can’t say it was any worse than the old Amazon “customers also bought” lists. But like those old lists, it’s also clearly algorithmic. An actual sci-fi fan would, I think, have a much better understanding of what someone was likely looking for when they asked that question. (At least if they had read the books involved.)

    Liked by 1 person

  4. Yeah, yuck indeed! And you’re right, it is market research. Seems like a lot of the business side of things is falling on authors nowadays. Not fun at all.

    I think it must be fairly easy for the AI to get the genre right, just because the publisher has to indicate that on various platforms and on the book itself. But I think you’re right, it can’t beat a knowledgable human. I’m not sure it can get into things like stylistic differences, though maybe themes.

    I was fairly impressed with the fake book descriptions, though. I think I would’ve used it to come up with some examples of phrases to use in writing my own query, though I’d have to get just a little more specific. The problem with writing these damned query letters and synopses is distilling an entire 90K or so word work that you’ve just been steeped in to a couple of snappy paragraphs. It’s hard to get general at that point, and hard to see your own writing objectively. The tendency is to include too many details. I ended up getting a lot of feedback like, “Tina, that’s not what your book’s about!” You’d think I’d know what my book is about, right?

    Liked by 1 person

    • The fake book descriptions could be useful as writing prompts. Although I don’t usually think of a novel as something done from a prompt.

      One of the benefits of outlining might be being able to come up with those synopses. Of course, that assumes the original outline remains a reasonable approximation of the finished work. And if a book is operating on multiple levels, one with the characters going to various places and doing various things, and another involving philosophical discovery, I can see coming up with a reasonably representative synopsis being a serious challenge.

      On knowing what your book is about, that reminds me of something an author once said on a podcast, that she never worries about the theme of her books. It comes out organically and is usually more apparent to others (notably the critics) than her. It’s a reminder that our writing often reveals more about ourselves and our state of mind than we realize.


      • Yeah, outlines would be a good start, but don’t forget, the whole point of the synopsis (if you’re asked for it) is to entice. So it can’t just be a rundown of plotting notes, or at least you still have to know what to keep and what to leave out. (That’s the hard part for me.)

        I feel somewhat the same as the podcast author you mention as far as worrying about the theme of my writing, but I do worry when I get to the editing stage (after draft one). That’s when I need to know what the theme is so I can reverse engineer things to make that theme stand out. But during that first draft I definitely take note of what my writing group notices, and yeah, it’s sometimes a surprise.


        • That makes sense on the synopsis. I imagine you have to know your audience. In this case, they’re not just interested in how much they’d enjoy reading the book, but probably the execution as well, in seeing that the author knows what they’re doing, and that the story has some form of satisfying structure. Although on that last point, given what I see in many books, I don’t know how much publishers scrutinize established authors on it.

          Liked by 1 person

  5. Reading your blog lately has been an eye-opener regarding the nitty gritty dirty details of book writing. I suppose no job is without its downsides, but you’ve certainly blown away the myth that writers just sit around and write. I guess you have to become famous enough to hire a flock of assistants to handle the crap jobs.

    I haven’t explored any of the LLMs but have seen a lot of demos both on the great things they can do as well as on their flaws and fragilities. They’re clearly very powerful tools, but it’s worrisome that we don’t yet understand how they “think”. Until we do, it seems unwise to let anything important depend on them.

    I can see them being useful for a writer. They’d be great at things such as: a list of fake book titles, some computer code, a letter written in a foreign language, part of a student essay, and a lot of other “boilerplate” that might be a pain for the writer.


    • God I am so late to the game. I had to look up LLM😆.

      I think you’re right about boilerplate. And that’s strangely hard for me to write; I’m better at going off on tangents and making sense of it all after the fact.

      On writers not sitting around writing, definitely. Some of the self published authors are quite business savvy, as they have to be. I wonder how they find time to write when they’re doing such a great job of selling, especially since those authors tend to be prolific.

      Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.