Write Useful Books: A modern approach to designing and refining recommendable nonfiction

by Rob Fitzpatrick

For advice about writing the prose itself, I recommend On Writing Well, by William Zinsser.

A major theme of this guide is to stop writing your manuscript in secret and start exposing it to — and learning from — real readers as quickly as possible. That might feel scary, but there are ways to do it safely, and it’s worth doing. You want to find (and fix) your book’s mistakes before launch, not after.

From our perspective as authors, this imbalance in recommendability is the million dollar mystery, and unravelling it is our goal throughout this chapter. Useful books are problem-solving products You can divide nonfiction books into two categories by their purpose to the reader: Pleasure-givers (“interesting”, “fascinating”, “beautiful”) Problem-solvers (“useful”, “actionable”, “clarifying”) Pleasure-givers are crafted like art or literature, as a solitary act of genius. Whereas problem-solvers ought to be designed and built like products, through a reader-centric process of testing and refinement.

Make a clear promise and put it on the cover Here’s the secret to a five-star Amazon rating: be clear enough about what your book is promising that people can decide they don’t need it. It may seem counterintuitive to try to drive potential readers away. But good books receive bad reviews after making too broad of a promise and luring the wrong people into buying. You can’t fully prevent bad reviews from ever happening, but you can certainly make them a rare exception by plainly stating who your book is for and what they’re going to get out of reading it.

When someone asks what you’re working on, attempt to describe the book in just one or two sentences. And then you need to do the hardest thing of all: to shut up and listen to them completely misinterpret and misunderstand what you’re trying to do.

Designing the world’s best TV for everyone is impossible. How could you possibly hope to figure out the right price point or feature set? Whereas designing the world’s best TV for a specific type of someone — the hearing impaired, parents of young kids, executive boardrooms — becomes instantly tractable.

Nearly every author attempts to include too much stuff for too many different types of readers. But that’s a recipe for writing something mediocre for everybody and mind-blowing for nobody — every chapter that the amateur adores, the expert endures, and vice versa.

The scope of a useful book is like the executive summary of a new business. It’s an as-brief-as-possible description of what it is, who it’s for, and why they’ll pay for it: Scope = Promise + Reader profile + Who it’s not for + What it won’t cover (And we’ll soon add recommendability and longevity as the final two considerations of a strong scope.)

What does my ideal reader already know and believe?

When someone decides to buy and read your book, what are they trying to achieve or accomplish with it? Why are they bothering? After finishing it, what’s different in their life, work, or worldview? That’s your book’s promise. What does your ideal reader already know and believe? If they already believe in the importance of your topic, then you can skip (or hugely reduce) the sections attempting to convince them of its worth. Or if they already know the basics, then you can skip those. Who is your book not for and what is it not doing? If you aren’t clear on who you’re leaving out, then you’ll end up writing yourself into rabbit holes, wasting time on narrow topics that only a small subset of your readers actually care about. Deciding who it isn’t for will allow you to clip those tangential branches.

Is your book’s promise Desirable enough that people will readily complain about, receive advice, give advice, and search for solutions to it? When someone encounters this problem/question/goal, is finding a solution a top priority or simply a nice-to-have? If your book could have several possible promises, does one have more “hair on fire” urgency for a certain type of reader? Of your several potential reader profiles, does one more actively search for (or give) advice and recommendations? Do any feel the pain more sharply? If so, they’ll fuel a stronger, faster recommendation loop.

Pick a promise that will remain relevant and important for 5+ years Avoid overreliance on temporary tools, trends, and tactics that are likely to become quickly dated

To see why this all matters, consider The 4-Hour Workweek (2007), by Tim Ferriss, a category-breaking bestseller that soothed a painful emotional “problem” around disillusionment with the 9-to-5. The book enjoyed incredible success due to both its own strengths plus Ferriss’s incomparable skill and hustle as a marketer. And yet, rereading it now, a dozen years later, a significant percentage of its content feels — at least to me — irrelevant and dated. Of course, huge amounts of value still exist! The good stuff is just hidden between extended discussions of tools and tactics that haven’t aged quite so well. And when only a portion of the book remains relevant, the whole is diminished.

To create a book that lasts and grows, the formula is simple: do the best job of solving an important problem for a reader who cares, without anchoring yourself to temporary tools, tactics, or trends.

Once you begin to feel that your book’s scope is resonating with your future readers, it’s time to shift toward testing the Effectiveness of what it will contain. Which begins by writing down a first guess at a detailed table of contents. Fill your ToC with takeaways, not clickbait Your ToC is the blueprint of your book’s education design. To serve its purpose as a tool for design and feedback, it must be built from: Clear, descriptive language Detailed subsections.

Hemingway quipped that you ought to “write drunk, edit sober.”

Stop writing for an anonymous crowd; imagine yourself writing to a specific individual who you know, and who wants your help.

Although a few rare authors appear to thrive on an impulsive schedule, the more common approach is to pick a time when you’ll sit at your desk and then sit there, religiously, even if you aren’t getting anything done. You don’t have to write, but you aren’t allowed to do anything else. Eventually, the boredom outweighs the writer’s block and you start typing.[10]

I’ve learned that if I don’t do my writing first thing, then I don’t do my writing.

adding word counts to the titles of your sections and chapters, allowing you to see how many words (and thus how many minutes — 250 words per minute is typical) are sitting between any two pieces of value.

Given the typical reading speed of 250 words per minute, cutting 10,000 words (while maintaining the value) saves 40 minutes of your reader’s time.

Front-load the value The likelihood of your readers recommending your book is based on the amount of value they’ve received before either finishing or abandoning it. And they’re most likely to abandon at the start.

Nonfiction authors make this mistake all the time via the inclusion of lengthy forewords, introductions, theoretical foundations, and other speed bumps that come from a place of author ego instead of reader empathy.

Revisions are not copy edits; they are major surgery and they suck. I didn’t understand this one for the longest time. I would hork up a first draft, turn back to page one, and start hunting for typos, feeling smug. Don’t do that. That’s like digging up a lump of coal and spit polishing it in front of Tiffany’s. … Don’t read it like it’s your precious perfect baby darling. Read it like it’s your worst enemy’s magnum opus and your job is to expose its every tragic flaw. … If you get bored reading it, so will your audience.[18]

If you’re not embarrassed by the first version of your product, you’ve launched too late. Although Reid’s advice doesn’t quite apply to the final launch of a finished book (since it’s relatively difficult to update once published), it applies 100% to exposing the pre-launch manuscript to beta readers. If you’re completely proud of it, then you’ve waited too long.

Three strong signals that your manuscript is “finished” and ready to be Polished: It feels easy to recruit new beta readers, since they want what you’re offering (Desirable) Most of them are receiving the value and reaching the end (Effective and Engaging) At least some of them are bringing their friends (the recommendation loop is running)

Google Docs is free and works well enough for this task. Set the permissions to allow people to comment/suggest, but not edit, and optionally disable their ability to download or duplicate the document. You can either invite them individually (for tighter access controls), or simply make the manuscript public-by-default and then share the link.

Alternatively, Devin and I have built a tool specifically for better beta reading called Help This Book,[19] which is designed to gather more (and better) data as well as to help you make better sense of it. I used it for this book:

You also need to tell your beta readers what type of feedback you need. Otherwise, they’ll spend all their time flagging typos when you’re stil...

Devin and I messed up basically everything about the launch of The Workshop Survival Guide (more on that in Chapter 7). But one thing we did right was to get signed copies into the hands of fifty influential review readers. The logistics took some time, but it wasn’t difficult — we just found their emails, said we wanted to send them a new book, and asked for their postal addresses. We asked for nothing in return, but some of them went out of their way to help us anyway.

In the early versions of The Mom Test, I had included a silly mock conversation to demonstrate how getting feedback tends to go wrong. I felt a bit sheepish about its goofiness and had only used one such example. But beta readers loved it, saying that it solidified a normally abstract concept.

Don’t worry too much about readers who want to find a way to argue about every tangential point. … Just enjoy writing.[20]

One bit of advice I give writers is to see each draft as a hypothesis or experiment: your job is to gather data to test that version of the manuscript and figure out what’s wrong with it. If it fails, it doesn’t mean that you have failed, but only that the current experiment has. So you redesign it. Shift your emphasis off the personal and back toward the product. Perspective is everything!

The problem isn’t you. The problem is the problem. Pressfield’s meaning here, at least as I understand it, is that it’s never you against your readers. It’s you and your readers working together against the problems in the manuscript.

If you ever become tempted to argue with a beta reader, do yourself a favor and step away. Get a good night’s sleep or take a week’s vacation. Attempting to “win” in a comment thread is clear evidence that you’ve started taking things a bit too personally and could use some time away.

The challenge with detecting boredom is that most readers are too polite to explicitly mention it. Or they won’t even notice what’s happening, blaming themselves for “not paying attention” or “feeling tired” when the problem is actually a failure of reader experience design.

This is yet another reason to invite actual beta readers instead of just pressuring your friends or hiring a professional. Friends and professionals feel obligated to finish the manuscript, which denies you the invaluable data of where they’re getting bored and wanting to give up.

Launch is a year, not a day.[22]

My top four suggestions for seed marketing (in no particular order) are: Digital book tour via podcasts and online events (most scalable) Amazon PPC (pay-per-click) advertising (easiest but unscalable) Event giveaways and bulk sales (fastest if you have the contacts) Build a small author platform via content marketing and “writing in public” (most reliable and valuable, but time-intensive)

If I were starting over with absolutely zero resources, reputation, or connections, I would rely mainly on the fourth option — writing in public to build a small author platform — complemented by Amazon PPC ads. But since you’re probably not starting completely from scratch, you’ll want to be familiar with all four approaches, since one or another might offer an easier win.

If you’re not a natural with public speaking and interviews, compensate with extra prep. Here’s Derek Sivers again (who appeared in Chapter 4 with his reader-friendly author bio), on how he excels — against his natural inclinations — as a podcast guest: I’m a disappointing person to try to debate or attack. I just have nothing to say in the moment, except maybe, “Good point.” Then a few days later, after thinking about it a lot, I have a response. … I’ll tell you a secret. When someone wants to interview me for their show, I ask them to send me some questions a week in advance. I spend hours writing down answers from different perspectives, before choosing the most interesting one. Then when we’re in a live conversation, I try to make my answers sound spontaneous. People say that your first reaction is the most honest, but I disagree. Your first reaction is usually outdated. Either it’s an answer you came up with long ago and now use instead of thinking, or it’s a knee-jerk emotional response to something in your past.[25]

Interestingly, PPC (“pay per click”) ads work better for self-published authors than those who are published traditionally. This is because a self-published author receives higher royalties per copy sold, allowing them to pay higher per-click prices while still turning a safe profit. (This is also one of the reasons to resist pricing your book too cheaply — it hinders your ad campaigns.)

If you’re familiar with other advertising platforms, you may be surprised to learn that you’re largely powerless to customize what appears in your Amazon ads. Your book’s cover and title/subtitle are its advertisement — you can’t just add a compelling photo or catchy tagline to make it more clickable. So if your cover is illegible and your title/subtitle mysterious, then your ads won’t work.

To find willing events, you must first understand that from their perspective, even a “free” book is never free — any sort of giveaway carries a high reputational risk by acting as a tacit endorsement. So the organizer must believe, first and foremost, that the book is useful to their attendees.

One slightly sneaky trick is to research relevant events while your book is still being written, and coax a few organizers into becoming beta readers. This is a smaller request (asking for their time instead of their reputation) that allows you to begin demonstrating value and building a relationship.

To do a physical giveaway, you’ll need physical books. However you decide to handle the printing,[28] a typical paperback will end up costing you roughly $2-5 per book, depending on the number of copies ordered and the page count of each. However, you’ll rarely need to pay for printing out of pocket. Once an event has decided to bear the reputational risk of ...

If you’re in the business of paid consulting or speaking, you can easily upsell clients by adding their logo to the cover an...

You can write your book in public, chapter by chapter or section by section, and just continually release these things to an ever-growing audience of people. Nobody will compile it into a book and release it without you. Doing that consistently, every single week, will build an audience whether you like it or not.[29]

Once a day, after you’ve done your day’s work, find one little piece of your process that you can share. If you’re in the very early stages, share your influences and what’s inspiring you. If you’re in the middle of executing a project, write about your methods or share work-in-progress. If you’ve just completed a project, show the final product, share scraps from the cutting-room floor, or write about what you learned.