Expert and business books

The funnel around a book, built honestly

August 16, 2026 · The StoryProp team

Someone has your book open on a train, or on the couch at the end of a long day, and reaches the final paragraph of the final chapter. For the length of that page they are more persuaded by you than they will ever be again — your vocabulary in their head, your argument freshly agreed with. Then they close the book, and nothing happens next, because nothing was built to happen. That silence is expensive, and it is the whole case for building a funnel honestly: the next step after the last page, designed rather than left to chance. In the Business Book ROI Study, a 2024 survey of 301 published nonfiction authors led by Josh Bernoff and paid for by four author-service vendors, 47% of the revenue authors attributed to their book came from something other than the book — speaking, consulting, workshops — and in Bernoff's follow-up analysis of the same data that share rises to 63% among authors who generated $10,000 or more and 69% among those who generated $25,000 or more. Nearly all of it happens after the last page.

The word itself makes serious writers flinch: countdown clocks, webinars that are really sales calls in costume. Keep the flinch — it is an instinct about respect, and it is right about the tactics it recoils from. But notice its cost. An author who refuses to think about the system around a book ends up with one anyway: a last page that stops, a copyright notice, a reader full of warm intention with nowhere to put it. That is a design too. It is just an accidental one.

What sits on the other side of the last page

Strip the jargon and a book funnel is only the ordered set of things a reader can do after the book, plus how they learn those things exist. By that definition every book has one. The default is your name in a search box: one unlabeled door, which works for readers already determined enough to go looking — the ones you were least likely to lose. Everyone else drifts off. Turning expertise into a book is a decision worth making deliberately, and so is everything after the last page.

The useful reframe is hospitality rather than acquisition. A guest who has been in your house for hours should not have to guess where the door is, and a reader who just finished is asking a plain question: I found this useful, now what. The next step is your answer. It might be a tool that makes the method usable on Monday, a place to ask a question, a workshop, a way to hire you outright. All legitimate. The design question is whether each door delivers what its sign promised.

Signpost or trapdoor

The difference between an honest next step and a manipulative one is not tone, and not whether money eventually changes hands. It is whether the reader gets what the sign said. The trapdoors are worth naming, because they have started to feel normal: withholding part of the book itself — the last chapter, the templates the chapters keep referencing — so the reader must trade something to finish what they already bought; deadlines that exist only in the copy; a closing page promising the material that would not fit, which turns out to be a discount code with a header on it.

Each does more damage than it looks. A book earns credibility slowly, across hours of attention, by being generous in a way that costs the author something. A bait-and-switch at the end spends that credibility in one click, and spends it backwards: the reader now rereads the generosity as technique. There is also an uncomfortable fact about who these tactics work on: the readers least practiced at recognizing them. That is a reason to refuse them well beyond squeamishness.

The extreme version of the trade is already on the public record, priced. The New York Times introduced its dagger symbol in 1995 to flag titles whose ranking was materially helped by bulk orders, after Bloomberg Businessweek reported allegations that a business book — Treacy and Wiersema's The Discipline of Market Leaders — had been bought onto the list; the authors and their publisher denied trying to manipulate sales. The undisputed case came later. In 2011 Mars Hill Church paid the vendor ResultSource $200,000 to place Mark Driscoll's Real Marriage at number one, which bought a single week on the advice list and a permanent liability the church itself later described as “manipulating a book sales system.” Two hundred thousand dollars for a week, and the week is not what anyone remembers. Back-matter trapdoors are the cheap version of the same purchase, and nobody writes those up.

Capture that doesn't feel like a trap

Capture is a hostile word for something that can be perfectly civil: an exchange, legible from both sides. When a reader hesitates at a form, the question is rarely whether an email address is a fair price. It is what this is going to turn into. Answer that where they can see it — what arrives now, what arrives later and how often, how to stop — in a sentence, in the book's own voice, without the coy vagueness that makes people assume the worst.

What sits behind the form has to be worth the exchange on its own terms. The strongest version is the tool form of the book's argument: whatever the chapters teach a reader to think, the resource helps them do. A book arguing that failures start in the reporting lines becomes an interview script for finding them; a book with a diagnostic at its center becomes that diagnostic as a working sheet. Built that way it is useless as generic bait, which is the point. And one good door beats an assortment — a reader facing a menu of downloads takes none, because choosing costs more attention than they have left.

Placement matters as much as the offer. Put the door where the appetite appears — at the end of the chapter that created it, in that chapter's terms — with a resources page in the back for whoever goes looking later. Design for the physical situation: a person holding a printed book retypes an address by hand, one-handed. Every extra character is a place to give up, and the page they land on must be unmistakably the right one.

Sequence by intent, not by ladder

Growth advice pictures a ladder — small thing, medium thing, expensive thing, climbed in order from the bottom rung. Readers do not behave that way. They finish with an intent already formed, and it is roughly one of three: keep learning, because the argument was interesting and the problem is not urgent; apply it themselves, because they have the problem and the appetite to do the work; or hand it over, because they have the problem and would like it to stop being theirs.

So the design is a set of clearly marked doors open at once, not a corridor. A reader who closed the book ready to hire you should not have to spend weeks in a nurture sequence to find out how, and a merely curious reader met with a proposal reads it as a misjudgment of who they are. Sequencing does have a place inside an intent — someone who took the worksheet has told you something, and the sensible next offer is help using it — while the slow channel that keeps a relationship alive as intent changes is a craft of its own. The three doors are not worth the same, either, and the ROI study puts a scale on the gap: counting only authors who reported income in a given stream, the median increase they attributed to the book was $50,000 for consulting, $40,000 for workshops and $30,000 for speaking fees, against $3,300 for direct book sales and $2,000 in royalties.

Those are conditional medians from a self-selected sample rather than a forecast, and the sample has a known tilt: respondents were recruited through the four author-service vendors who sponsored the study — a hybrid publisher, a ghostwriting agency, a PR firm and a thought-leadership consultancy — so it over-represents authors who invested in paid services, though the report states the sponsors did not influence or review the findings. Even inside it, 31% of the authors surveyed reported no revenue increase in any category. The volume figures are the more useful ones for design: more than 70% of authors reported an increase in leads or gigs after publishing, and among those who did, the study describes typical magnitudes of ten speaking gigs, seven workshop gigs, dozens of consulting leads or scores of coaching sessions. All of that is author-attributed rather than measured conversion, but ten gigs is the order of magnitude a last page is competing for, and it is not the sort of thing that arrives when the last page simply stops.

Counting what the book actually did

The same study found the thing that should change what you count: book sales did not predict ROI or goal achievement. Copies are the number everyone watches and the wrong one to steer by, partly because the lag between reading and needing runs to years, copies get passed along, and the reader is often not the buyer or the person who eventually signs anything. So the instrument worth building is the one most authors skip: a question in your intake — how did you come across us — and the habit of writing the answer down verbatim. Those sentences tell you more than a click path can. Which chapter people mention. Which line they quote back, rarely the one you would have guessed.

Scale is worth reasoning about in the open, with the real order of magnitude in view. Authors in the ROI study reported selling a median of 4,600 copies when traditionally published, 1,600 hybrid and 700 self-published — self-reported figures from a self-selected sample, not market measurement, and well short of the 8,900, 4,400 and 1,100 the same authors had expected. Those are your denominators, and they run through a chain. How many readers reach the part of the book where the door is: attention. How many notice and understand it: placement and clarity. How many find the step worth taking: fit between the resource and the argument. How many have the problem live right now: timing you do not control. The links multiply, so one weak link makes everything after it look broken — which is exactly why placement and clarity carry so much weight when the starting number is four figures, not six.

The trap in measuring is that the metrics easiest to move are the ones manipulation moves. A more alarming subject line, an invented deadline, a headline that overpromises: each improves something visible while degrading something that is not, and the invisible thing is the slow one that produces actual work years later. A serviceable rule — refuse any tactic you would be embarrassed to describe accurately to the reader it was aimed at.

The last page is writing, not marketing operations

The closing note is the highest-attention real estate you will ever own, and it is usually the last thing made: assembled in a hurry at the end of an exhausting project, in a register borrowed from advertising. Readers feel the seam. A book that has been careful and specific for hours suddenly clears its throat and starts talking like a landing page, and the shift does retroactive damage to everything before it. Read the closing note aloud straight after the final paragraph of the last chapter and listen for the join.

That is why the funnel belongs in the drafting rather than in a spreadsheet afterward. The resources appendix, the closing note, the honest description of what a workshop covers and who it is wrong for — all of it is writing, subject to revision like any chapter. AI book writing software carries back matter as readily as chapters, holding the book's terms and voice steady through the last page, and the file you export carries your byline and nobody else's branding.

One test covers most of this. If a reader took every step you offered and never paid you anything, would they still be better off for having taken them? Answer yes and you have built hospitality, which is also what works: in the ROI study, 68% of authors said the book made them significantly more credible to prospects and clients, 59% saw more podcast interview requests, and 72% strongly agreed that writing it had been a good decision. That is what a well-marked last page compounds. Build the doors with the care you gave chapter three, and the book keeps working long after the reader closes it.

The back matter is where a book most often loses the reader it just spent hours earning, and it is written, not configured. StoryProp drafts the closing note and the resources appendix in the same voice as chapter one, with the terms you settled on kept on record, so the last page still sounds like the person who wrote the first.

← All posts