The First 10%: Writing a Look Inside Sample That Converts

Every book page has two distinct jobs. The cover and the description earn the click. The Look Inside sample — Amazon's free preview of roughly the first 10% of your ebook file — earns the sale. That slice of prose is the only place a stranger encounters your actual writing instead of your marketing copy about it, and serious shoppers use it before they buy. The implication is direct: the sample is not a courtesy feature. It is the audition. Listings with Look Inside enabled convert about 7% better on average than listings without it; a book with no preview reads, to experienced buyers, as a book with something to hide.

What does the Look Inside sample actually show — and how does front matter destroy it?

The sample is fixed at approximately 10% of the total ebook file by Amazon — not 10% of the story, not 10% of the narrative content. It is calculated from the byte size of the uploaded file, and it always begins from the very first byte, regardless of where you have coded the book to open on a reader's device. Every page that sits before your first chapter — the copyright notice, the dedication, the epigraph, the cast of characters, the "also by" list, the table of contents — counts against that 10%, and the sample cannot skip past it to reach Chapter 1.

The failure mode is vivid. One author reported in a 2024 community survey that downloading a Kindle sample yielded, after front matter ran out the window, roughly one page of actual prose text. That is the entire audition, spent on legal boilerplate and a dedication nobody browsed the book to read. The story never appeared in the sample at all.

The fix is unglamorous and nearly free. For a Kindle ebook, only two front matter elements genuinely belong before the story: a title page and a copyright page (one to three lines of essential legal text). Everything else belongs in the back of the book:

Nonfiction gets one clear exception: keep the table of contents up front. Nonfiction browsers treat it as a purchasing tool, scanning it first to confirm the book covers the ground they came for, and it earns its place in the sample window. A fiction table of contents listing only "Chapter One, Chapter Two" carries no equivalent value and burns sample space the story should occupy. Strip it or move it.

Once front matter is trimmed, verify the result in Kindle Previewer — Amazon's free formatting tool — on a phone, because the large majority of Amazon book purchases happen on mobile as of 2026. A sample that opens on a copyright page on a four-inch screen costs you every shopper who reads that way.

What makes a first line and first chapter actually win the browser?

The attention window that greets your opening is genuinely narrow. Average reader attention fell from 150 seconds in 2003 to approximately 47 seconds in recent years — a 69% decline. And 15.8% of readers abandon a book within the first 50 pages, per ProWritingAid's study of 125 years of bestselling first chapters. You are not being read by a patient person in a quiet room. You are being skimmed by someone with a thumb already drifting toward the back button, deciding in under a minute whether your prose is worth their time.

K.M. Weiland's formulation is the right anchor: "stripped to its lowest common denominator, the hook is nothing more or less than a question." The first line must raise at least one unanswered question the book actually goes on to answer. The named hook patterns that do this fastest:

• The disturbance. Drop a disruption — a phone call, a body, a gunshot — into the opening line and reveal its consequences before any backstory. James M. Cain's "They threw me off the hay truck about noon" delivers character, conflict, and voice in one sentence, with zero context required of the reader.
• The absence hook. Open on what is not there — a cold half of the bed, an empty chair — so loss and stakes register before anything overt happens. Suzanne Collins opens The Hunger Games with "When I wake up, the other side of the bed is cold" and breaks the common "no waking-up openings" rule precisely because the cold half communicates dread without stating it.
• The intriguing statement. A flat declarative that inverts an assumption. "It was a pleasure to burn" forces the reader forward to understand how that can possibly be true. The statement should challenge what the reader assumed was obvious fact.
• The voice hook. A line so distinctly narrated it could only come from this character, making the personality itself the invitation. The line tells the reader: there is a mind here worth spending time with.
• The problem-first hook (nonfiction). Name the reader's specific, felt failure before offering the system that addresses it. Validate the frustration, then promise the fix. This is the move James Clear makes in Atomic Habits: "You do not rise to the level of your goals. You fall to the level of your systems."
• The micro-to-macro anecdote (nonfiction). One specific person in one specific moment, just vivid enough to believe, then the zoom out to the pattern that governs thousands — the dominant opening structure in big-idea nonfiction bestsellers.

The ProWritingAid data gives the quantitative shape of commercial openings: emotion-tell rate averages only about 1% in top-performing first chapters, meaning those books show a character's internal state through action, dialogue, and thought, almost never by naming the emotion directly. Passive voice averages roughly 5% of sentences. Slow-paced paragraphs average 14%, with BookTok titles running as low as 9%, and romance — the fastest-paced genre in opening chapters — averaging only 7%. These are snapshots of central tendencies in commercial fiction, not absolute laws. A brilliant literary opener can break every one of them and still convert the right reader. But they describe the shape of what converts for most commercial books: lean, fast, dramatized prose that shows rather than explains.

One rule that holds across every genre: introduce the protagonist first. Readers bond with whoever appears on the first page, and a vivid stranger who vanishes by Chapter 2 reads as bait-and-switch. Whichever hook type you use, the protagonist should be present, and their external goal and internal stake should be visible within the first chapter — what does the character want, and what do they risk losing by wanting it?

Does the opening work differently for fiction versus nonfiction?

The governing principle is shared: open on the place where the purchase decision is made. The execution splits cleanly along genre lines, and confusing the two is among the most common reasons a sample fails to convert.

For fiction, begin at the moment the ordinary world is about to change — not the morning before it, not two chapters before it. The single most documented opening failure is a character waking up and describing their bedroom; the second is two pages of weather, landscape, or backstory before anything happens. Neither provides a story engine, and neither earns the purchase. The inciting incident — the disruption that launches the actual plot — should appear no later than the 10-to-15% mark of the total book, and often earlier. K.M. Weiland's structural model places the formal inciting event at approximately the 12% mark; Collins fires it in The Hunger Games at approximately 5% — page 22 of 374. The sample's job is to reach it while the preview is still free. A browser who finishes the free preview with nothing yet at stake has no reason to buy.

For nonfiction, the move inverts. Open on the reader's problem, not your credentials. The first question a nonfiction browser is asking — consciously or not — is: "What does this author have to teach me, and do I believe they can deliver it?" The fastest way to lose them is to spend page one on biography or publication history. State the wound first, promise the cure, then earn it. A secondary mechanical advantage: roughly two-thirds of nonfiction buyers scan rather than read the sample in full, which makes short paragraphs, bold lead-ins, and a transformation promise stated in the first paragraph conversion mechanics, not stylistic choices. A preface that buries its core promise on page three will not be seen by most browsers.

Why does the 15% Kindle Unlimited threshold make your opening do double duty?

Kindle Unlimited authors face a version of the sample problem that has a direct monetary translation. In KU, authors are paid per page read through the Kindle Edition Normalized Pages system — at a rate that averaged approximately $0.0043–$0.005 per normalized page in 2025, per Automateed's KENP payout tracker — not per borrow. A KU download that stalls at Chapter 1 pays nothing, because the page-read counter advances only as someone actually turns pages. Hugh Howey's reframe of this mechanic is precise: you lose money for every page left unread. Every craft decision that touches pacing is also a revenue decision.

Author-side analytics suggest that KU readers who make it past roughly 15% of a book tend to finish it; the primary early-abandonment drop-off is concentrated in that opening window. Serial fiction platforms record roughly a 50% reader drop between Chapter 1 and Chapter 2, then about 10% additional loss per chapter thereafter — and authors report meaningful completion-rate improvements from focused rewrites of the first two chapters. Top-performing stories achieve 80–90% Chapter 1 retention. The direction is consistent across data sources: the opening is where readers decide whether to stay.

Because the 10% Look Inside sample and the 15% KU completion threshold overlap almost entirely, a well-engineered opening solves both problems simultaneously: it converts the browser to a buyer and pulls the KU borrower past the abandonment zone into the completing majority. Shawn Coyne's Hook-Build-Payoff framework from Story Grid applies directly at this scale: the opening chapters must hook at their entry, build complication across the early pages, and pay off the promise that made the reader begin. Structure the first three chapters as a unit, not as isolated pieces, and both the sample conversion problem and the KU completion problem resolve from the same craft investment.

How do you end the sample chapter so a browser cannot walk away?

The last line of Chapter 1 is the line that closes the sale. End it on a hook — an unanswered question, a raised threat, a charged emotion left unresolved — so reaching the end of the free sample feels like a loss the reader can fix only by buying. A chapter that ends with the protagonist content and the dramatic question answered gives the browser a clean place to stop. Never give them a clean place to stop.

The specific ending techniques that work: the question hook (leave one specific, answerable question open — plot, character, or consequence — and show the situation rather than announcing that something is uncertain); the escalation ending (introduce a new complication before the existing one resolves, remove a resource, or reveal that apparent progress was a setback); and the emotional peak cut (end at the moment of realization or impossible choice, before the character acts, forcing the reader to turn the page for the outcome). K.M. Weiland's formulation holds at the chapter level too: imply the tension, do not editorialize it.

There is a more advanced optimization some authors pursue: engineering the 10% file cutoff to land right after a cliffhanger, so the sample itself ends mid-tension. It is difficult to land precisely, because the cutoff is a function of total file size — and the ebook sample at 10% and the print sample at 20% will never end at the same narrative moment. Lean front matter gives you more control over where the narrative falls at the file's 10% mark, but do not over-invest in hitting it exactly. Make Chapter 1 end on a strong hook regardless of where the cutoff falls; any reader who reaches the end of the sample will reach it wanting more.

One technique every craft source converges on: write Chapter 1 last. The complete draft is the only thing that tells you what your book is genuinely about — its arc, its theme, its ending — and only then can you build an opening that promises exactly what the book delivers. Many writers begin two chapters before the story truly starts. To find out whether you have, read what is currently your Chapter 3 as a cold open and ask whether it is the stronger beginning. If it is, start there and redistribute what came before throughout the text. The first draft is how you discover the story. Chapter 1 is what you write for the reader once you finally know it.