Romance is the most structurally honest genre there is. Readers know the ending — together, happily — so the book is judged entirely on how it earns that ending, and the beats are how it earns it. This is a working romance beat sheet: each beat, roughly where it lands in an 80,000-word novel, what it has to accomplish, and the question a developmental editor actually asks when checking it. Copy it into your outline or run your finished draft against it.
There's no single official romance beat sheet. The structures most authors use descend from Gwen Hayes's Romancing the Beat, from Jessica Brody's Save the Cat! Writes a Novel adapted for romance, and from the beat sheets Jami Gold has published for years, all of which are worth reading in full. What follows is the version we check manuscripts against at Authors Starport, which folds those into one sheet and adds the editor's question at each beat. The percentages are positions, not rules — a beat that lands five thousand words early or late is fine; a beat that's missing is not.
| Beat | About where (80k) | What it has to do | What an editor checks |
|---|---|---|---|
| 1. Setup / ordinary world | 0–5% (0–4k) | Show each lead's life before the other, and the wound and lie that make love feel impossible right now. | Can I name the lie each lead believes about themselves in one sentence? If not, the ending has nothing to pay off. |
| 2. Meet-cute / the meet | 5–10% (4–8k) | The leads collide, and the collision shows the conflict between them in miniature. | Does the first scene together contain the whole book's friction? A pleasant meeting is a missed beat. |
| 3. No way / refusal | 10–15% | Each lead has a concrete reason this can't happen — external and internal. | Is the reason a real obstacle or a misunderstanding one honest conversation would fix? Readers forgive the first, not the second. |
| 4. Adhesion / forced proximity | 15–25% (12–20k) | Something locks them together — a job, a bet, a fake engagement, a snowed-in inn — so they can't walk away. | Is the adhesion strong enough to survive the midpoint? Weak glue is the number-one reason romance middles sag. |
| 5. Fun and games / falling | 25–50% | The promise of the premise: the banter, the near-misses, the tropes the reader bought the book for, and the first real cracks in each lead's lie. | Does every scene here change something between them, however slightly? Scenes that are only fun get cut in a developmental edit. |
| 6. Midpoint | 50% (40k) | A false high (first kiss, first night together, a declaration) or a false low — either way, the relationship shifts from reacting to choosing. | Does the midpoint raise the stakes and tighten the inner conflict? A first kiss that changes nothing is a scene, not a midpoint. |
| 7. Deepening / intimacy | 50–65% | They see each other clearly for the first time; vulnerability; the lie is openly challenged. | Has each lead said something here they couldn't have said in chapter one? That's the arc, on the page. |
| 8. Retreat / the shield goes up | 65–75% | The old wound reasserts itself; one or both pull back. The external plot squeezes at the same time. | Is the retreat caused by the lie established in beat 1? If it's caused by a new, convenient problem, the structure is broken. |
| 9. Black moment / break-up | 75–85% (60–68k) | The relationship ends, or appears to. Each lead's lie wins. It has to cost them something real. | Is this the inevitable result of the two lies colliding, and does each lead lose something they can't get back without changing? |
| 10. Dark night / the choice | 85–90% | Alone, each lead finally sees the lie for what it is — usually with help from a secondary character — and chooses to change. | Does the lead change themselves before the reunion? A reunion that arrives because the other person apologised is unearned. |
| 11. Grand gesture / declaration | 90–97% | The changed lead acts on the change, publicly or at cost — proof, not promise. | Does the gesture require the new self? If the chapter-one version of the character could have done it, it isn't a gesture. |
| 12. HEA / HFN | 97–100% | The happily-ever-after or happy-for-now, plus a glimpse of the life the two arcs made possible. | Does the ending answer the exact question the opening asked? The best last scenes mirror the first. |
The beats are the surface. Underneath, every romance runs two arcs at once — each lead moving from a lie to a truth — and the beats are simply the moments those arcs collide. This is why a beat sheet filled in mechanically still produces a flat book: the midpoint kiss is in the right place but neither character's lie was touched by it. When editors say a romance "hits the beats but doesn't land", this is almost always what they mean. If you want the arc side broken down, our character arc guide walks through the want, need, lie, midpoint turn and final choice that the beats above are built from.
For a word-count sanity check alongside this, see romance novel word count; for the middle specifically, how to fix pacing in a romance novel.
Tagging a whole manuscript against the sheet is a few hours of honest work. Authors Starport's Scene Map does the first half automatically — every scene in sequence, with what changes in each — so you can see where the beats are and where they aren't. The Developmental Critique does the second half: an editorial letter, written for romance structure, that names which beats are missing, late or unearned, cited to the chapter they live in. The free Reader-Ready Scorecard is the place to start.
Most writing tools give you an AI read of your pages and nothing else. Authors Starport is a full editing suite — developmental critique, a line edit delivered as tracked changes, a style sheet built from your own prose, a scene map and an instant score — and every one of those tools is grounded in what real readers actually respond to. Behind them sit 694,681 hook mentions extracted from real reader reviews, 137,874 book blurbs analysed for hook, stakes and tone, and 4,293 reader personas built from that data. On this site, since June 2026, authors have uploaded 74 manuscripts and run 214 completed critiques and edits plus 40 free Scorecards.
Compared with the beat sheets you'll find on craft blogs — which are excellent templates but can't read your book — Starport's Scene Map and Developmental Critique check your actual manuscript against these beats, with the judgement grounded in what real romance readers praise and reject.
Counted from our own databases on 21 August 2026. We update these figures when we re-count; we don't round them up.
The Reader-Ready Scorecard grades your full manuscript on nine craft metrics, benchmarks it against professionally-edited books, and tells you exactly what to fix first. Free, in minutes.
Get your free Reader-Ready Scorecard →In order: setup, the meet, refusal, adhesion (forced proximity), fun and games, midpoint, deepening, retreat, black moment, dark night and choice, grand gesture, and the happily-ever-after or happy-for-now. The percentages vary by author and by system, but the sequence is remarkably stable across the genre.
Around the 50% mark — roughly 40,000 words into an 80,000-word novel. It's usually a false high (first kiss, first night together, a declaration) or a false low, and its job is to shift the relationship from reacting to choosing while raising both the external stakes and the internal conflict.
The break-up, around 75–85% through the book: the moment the relationship ends, or seems to, because each lead's lie about themselves wins. It has to be the inevitable result of the conflict set up in the opening and it has to cost each lead something real — a break-up caused by a new, convenient misunderstanding is the most common structural fault editors flag.
No. The positions are guides, not rules; a beat that lands a few thousand words early or late is fine. What matters is that each beat exists, that it's caused by the beats before it, and that the two leads' inner arcs actually move at each one. A mechanically-filled sheet with untouched arcs produces a book that hits the beats and still feels flat.
Authors Starport's Scene Map lays out every scene in sequence with what changes in each, which is the fastest way to see where the beats are and aren't, and the Developmental Critique names missing, late or unearned beats in an editorial letter cited to your chapters. Both are grounded in what real romance readers respond to; the free Reader-Ready Scorecard is the starting point.