Review of ‘Write Useful Books’, by Rob Fitzpatrick
I paid $24 for a PDF of a manuscript with typos and “TODO”s in it, and it was totally worth it.
Write Useful Books may be the least abstract thing I've ever read. It contains sentences like:
If you sell a PDF [of your book] anywhere, expect it to be pirated. As such, treat it like grassroot advertising by creating a call to action inside the PDF to bring readers back to your newsletter, website, or other products.
The advance from a big publisher typically helps reveal how much you believe in your book, and thus their willingness to provide guaranteed marketing and promotion. If a big-name publisher offers an advance of $50k or more, you're probably on track to receive the best they have to offer.
Notice how it straightforwardly explains both the logic and the specifics? What is a ‘big’ advance? Apparently 50k (in the US in 2021).
It’s also delightfully self-demonstrating. As a useful book, it practices what it preaches. It tells you to make your subtitles sentences, and its subtitles are sentences. There’s a CTA for Fitzpatrick’s newsletter inside the PDF I got (bought!).
The meat of the book consists in (1) teaching you how to create a book that useful enough for people to recommend it to one another (growing its audience exponentially), and (2) teaching you how to create an initial 'seed audience' that can get the process going. (These are not distinct steps, as Fitzpatrick explains – building your seed audience doubles as recruiting early beta readers, who help you improve your manuscript until it is sufficiently useful.)
The key pints:
Figure out what clear, well-defined problem your book should solve for what kind of reader. Then solve it for that reader by defining concrete takeaways, insights, skills, or ‘a-hah’ moments that they were looking for. Put these on plain display as your table of contents.
Value-per-page is essential. People stop reading mostly out of boredom or because you are making them wait for the value payoff.
Get waves of reader feedback from beginning to end.
Book marketing and creating the book are deeply intertwined. Write in public to both build a seed audience and make your work a lot better.
A few more random, more specific pieces of advice I thought were, again, charmingly and usefully concrete:
Put the promise of your book on the cover in a way that will be legible in a thumbnail on Amazon.
“If you get stuck, try writing the book in your email client. Put one section's title in the subject line and address it to a friend who knows what's going on. And then, in the body, simply type out in the shortest note possible to explain or justify the subject line. That's the first draft of that section.”
Once you have a manuscript, find 3-5 beta readers. Redraft and find another 3-5. Do this over and over. Invite 4x the people you need each time. Put the stuff on a blog/newsletter/social media following/reddit/wherever, suggesting people ping you if they’re up for commenting. Actually use their feedback.
I took down a lot more great information from the book, but won’t write it here so as not to screw Fitzpatrick over with this review.
Minor critiques:
As you can tell I loved the book. But I thought I’d share a few nitpicks just to dignify this post a bit:
Fitzpatrick implies that if you write in his way – by using tons of reader conversations from the beginning and scoping the book properly – book-writing is no longer high risk. I agree it greatly reduces the risk of the book being a flop, but it probably increases the risk of the book
Taking it forever to finish (or not finishing at all).
Straying from its original goals, especially if it has an activist and therefore not necessarily reader-pleasing component (this isn’t always a bad thing).
Experiencing ‘scope creep’ – some readers will want more of various things and you’ll want to give it to them. In general, the more stakeholders, the more scope creep.
Maybe this will be counteracted by other readers getting board by all the extra stuff and giving you negative feedback on it.
I think Fitzpatrick underplays the costs of the reader conversations and constant feedback, though he is probably right about the huge benefits. E.g., I'd guess conducting the two initial types of reader conversations does not cost “hours,” as he says, but rather days or even weeks.
I wonder if the people who are most willing to give you feedback and have initial conversations with you about your book will be adversely selected or at least selected: will they be disproportionately likely to have a lot of strongly held opinions of varying quality and have a lot of time on their hands? It sort of makes me think of how all the comments on Hacker News are terrible. I probably wouldn’t want HN commenters as my dominant source of feedback.
Notable in this context that here I am giving feedback…
Anyway, I was very happy with Write Useful Books, a delightfully meta, extremely clear, and as far as I can tell generally right-headed manual for creating and marketing high quality, useful nonfiction.
I am also happy to be illustrating the fruits of its own points with this very recommendation.
