The Treatment Works in Prose but Breaks in Layout. What Needs Rewriting?
The Treatment Works in Prose but Breaks in Layout. What Needs Rewriting?
The prose document reads fine. You put it into pages and now something on the page is false.
The reflex is to cut words until it fits. That reflex assumes the problem is volume, and very often it isn't. What the arrangement destroyed is a relation: a cause that lost its consequence, a qualifier that now arrives after the claim it was meant to limit, a reference the reader meets before they know what it is for. Cutting copy treats a connection as though it were padding, and it can remove the one sentence holding the meaning together.
So the question isn't how much the page can hold. It's which relation broke, and which is the smallest repair that restores it — moving the information, rewriting the connection, or changing the sequence of images. Each of those has a cost. Name the cost before you choose.
The worked case below is invented: proposed page descriptions, not renders, with no reader tested on them. It exists to make the choices visible, not to report how people read.
Read the page as arranged and write down what it now asserts
The method is uncomfortable and quick. Read the spread in the order the layout creates — dominant image first, then headline, then body, then caption, then the turn into the next page — and write one sentence stating what a reader would conclude from exactly that, having read nothing else. Then compare the sentence with what the treatment agreed to say.
Here is the invented case. A fictional spot, "Third March," for a fictional product, Vinter Groundwork's soil conditioner. A gardener, Marta, works the conditioner into the top foot of her allotment bed in October, alongside the leaf mould she has kept for years. Two winters pass. In the third March, the bed that fought the fork in her first March turns over, and she is quietly pleased. Writer and designer agreed on that restraint: nothing transforms on screen, the conditioner is one contributor among several, and the interval is the whole point of the film.
Three images and four pieces of copy, used identically in both spreads below:
- Headline: "Plot 14."
- Caption one: "October. Marta works the conditioner into the top foot of plot 14, with the leaf mould she has kept for years."
- Caption two: "Two winters later."
- Caption three: "The bed that fought the fork in her first March turns over."
- I1: October wide, Marta working the bed.
- I2: the third March, the bed turning over, Marta crouched with a hand in the soil.
- I3: the pack on a bench, label legible.
Proposed sequence A, arranged so the interval is disclosed before its consequence. Left page: "Plot 14"; I1 dominant; caption one beneath it. Right page: I2 dominant; caption two directly beneath it; caption three below that; I3 small at the foot.
Reading order: the plot, October, the conditioner worked in with the leaf mould — then, over the turn, a bed turning over and a pleased gardener, with "Two winters later" attached to the moment before the sentence that describes it. The pack arrives last.
Proposed sequence B, same elements, different placement. Left page: "Plot 14"; I3 dominant; caption one beneath it; I1 below that. Right page: I2 dominant; caption three beneath it; caption two at the foot, smallest type on the spread.
Reading order: the sack, label legible — then October and the leaf mould — then the pleased gardener and the turned bed — and only then, in the smallest type on the spread, "Two winters later."
The sentence for B is something like: the conditioner turned the bed over, and it did so within a season. The interval is technically present and practically inert. It arrives after the consequence it was meant to qualify, so the causal claim has already been extracted by the time the reader reaches the correction. There's a second loss, separate from the misreading: the pack, the least story-rich image in the set, has taken the strongest position on the spread.
That's the diagnostic. It isn't "too much copy." Neither spread is denser than the other. Not one word differs between them.
A caution about A: it is the arrangement the writer and designer believe reads as agreed. Belief isn't a measurement, which is why the check near the end of this piece exists.
Placement problem, or two people wanting different films
Before repairing anything, ask whether the writer and the designer intend the same event and the same consequence.
In the invented case they do. Both want restraint; both agree the bed turns over in the third March and not before. What broke is the order in which a reader meets the parts. That's a placement problem, and placement problems are cheap.
The other possibility is that one of them wants the soil to lift in a sheet while the other wants a woman quietly crouched with dirt on her hands. If that's what's happening, no rearrangement will settle it, because the two pages are not failing to communicate one idea — they are communicating two. Moving a caption just changes which of them reads first. That decision is about the film, and it belongs upstream of the layout pass. Settling it silently in the pages is how a treatment gets rewritten by nobody in particular, and how the writers end up defending a film they didn't pitch.
A quick test: if moving a caption fixes it, it was placement. If moving the caption leaves the page still arguing with itself, the disagreement is about intention, and it needs a different repair than the one this piece is describing.
Compare the three repairs and what each one gives up
Three repairs are available in the invented case. They are not ranked by how few words they change.
Move the qualifier. Lift "Two winters later" from the foot of the right page and set it directly under I2. No words change and no images move. The cost is spatial: I2 had the right page to itself, and the caption now competes with it. If I2 runs full-bleed, the line has nowhere clean to live — it goes over the picture or the picture shrinks.
Rewrite the connection so it travels with the claim. Fold the interval into caption three — "Two winters on, the bed that fought the fork in her first March turns over" — and drop the standalone line. The qualifier now sits inside the sentence it governs, where it will survive the next rearrangement as well as this one. The cost is cadence. "Two winters later" as a line on its own is the film in four words; folding it into a longer sentence spends the beat it was providing.
Resequence the images. Take the pack off the left page and open on October; put I3 at the foot of the right page, after the interval and after the pleasure. This is the largest change and the one with the clearest non-logical cost: the client's pack is no longer the first thing anyone sees. It also asks the October wide to work at page scale, which is a different picture problem than the one you were solving.
Sequence A is the first and third repairs combined. It is one solution, not the answer. The second repair would work on its own if the pack has to lead, and the first alone may be enough if the right page can absorb the caption.
The criterion for choosing is not which repair is shortest. It's which loss the film can spare. Losing the standalone beat, losing the pack's prominence, losing the reaction's scale — pick one, and say out loud which one you picked.
Why cutting the qualifier isn't one of the three
The tempting move is to delete "Two winters later" and let the page breathe. It saves a line and requires no conversation with the designer, which is exactly why it gets done.
It also changes the claim. Without the interval, the spread reads: she dug the conditioner in October, the bed turns over, she's pleased. That's a single-season transformation. The film was about how long this takes. Nobody had to agree to remove the argument; it left with the words.
The neighbouring cut is "with the leaf mould she has kept for years." Also a line saved, also a different claim: the product did this alone. The treatment said it was one contributor of several.
Then there's the quieter version. Keep the qualifier, set it at six point at the foot. Nothing has been deleted, so the page passes an inspection that counts words. Whether that line still lands inside the path most readers take through the spread is a judgment about reading paths — and the honest form of that judgment isn't an assertion about how eyes move. It's the check in the next section.
The general point: smaller type doesn't reconnect a broken relation. It makes a misplaced sentence slightly harder to find. But the reverse matters too, and it's the part people resist. Not every line is load-bearing. If the caption under the pack repeats October and the leaf mould that caption one already established, that repetition can go and nothing is lost. The test is whether the words carry a relation that nothing else carries.
Read the repaired order, at size, with the neighbours
Assemble the export at the size it will actually be viewed — not a zoomed-in artboard where every caption is comfortably legible — and include the page before and the page after, because the turn into the spread is part of how the spread is read.
Then ask someone who hasn't heard the pitch to say, in their own words, what happens in the story and what still looks conditional. Ask before explaining anything.
What you're listening for is whether the interval survives the telling. Do they say "over two winters," or "then she used it"? Do they describe the bed as turned over or as transformed? Those are different films, and the difference shows up in a sentence, not in a rating.
Treat what they say as one account of the pages you showed them, in the order you showed them, not as a rule about readers. If the work will be judged by a group, it's worth doing twice with different people; it's still an observation rather than a law. What it can do is tell you whether the small repair was enough or whether the pack has to move as well.
Then return the accepted change to copy and layout, and leave the rest alone. If the spread now reads as agreed, the sections that were already reading correctly need no attention. Reopening the whole film's voice because one spread misread isn't a repair, it's a restart.
What the fix looks like
In the invented case, the smallest repair is the first: move "Two winters later" to sit under the pleasure it belongs to, leaving the pack where it is. Three words relocated. What changes is not how much the page says but when the reader learns it.
The trade is real and worth saying plainly. I2 gives up the right page to itself, and the foot of the spread gets thinner. If that cost turns out to be worse at viewing size than it sounds on paper, the pack moves next — which is the third repair, and which asks the client to give up the lead image.
The second repair is the one I'd hold in reserve. Folding the interval into the sentence it governs is more durable than any placement fix, because it survives the next rearrangement. It also spends the beat the film is named for, and that's too high a price for a spread that will otherwise hold. Keep it for the pass that rebuilds the pages anyway.
None of that requires reopening the treatment's voice. It requires knowing which relationship the arrangement broke — a smaller question than it first appears, and a more consequential one than the page count suggests.
Frequently asked questions
When a treatment reads well in prose but breaks in layout, why isn’t cutting words the first fix?
The arrangement may have destroyed a relation: a cause that lost its consequence, a qualifier arriving after the claim it limits, or a reference the reader meets before knowing its purpose. Cutting copy can treat a connection as padding and remove the sentence holding the meaning together. Ask which relation broke and which smallest repair restores it—moving information, rewriting the connection, or changing the sequence of images—then name the cost.
How can I diagnose what a spread now asserts?
Read the spread in the order the layout creates—dominant image, headline, body, caption, then the turn into the next page—and write one sentence stating what a reader would conclude from exactly that, having read nothing else. Compare it with what the treatment agreed to say. If moving a caption fixes it, it was placement; if the page still argues with itself, the disagreement is about intention and belongs upstream of the layout pass.
What are the three repairs in the worked case, and what does each cost?
Move the qualifier: no words or images change, but the caption competes with the image and may have nowhere clean to live. Rewrite the connection so the interval travels inside the claim: more durable, but it spends the standalone beat. Resequence the images: largest change; the client’s pack no longer leads, and the October wide must work at page scale. Choose by which loss the film can spare.
Why not just delete the qualifier or set it in tiny type?
Deleting the interval changes the claim to a single-season transformation; deleting the leaf-mould detail makes the product seem to have done it alone. Tiny type does not reconnect a broken relation; it makes a misplaced sentence harder to find. Still, not every line is load-bearing—if a caption repeats what another already established, that repetition can go.
How should I check the repaired order?
Assemble the export at the size it will actually be viewed, including the page before and after. Ask someone who hasn’t heard the pitch to say, in their own words, what happens and what still looks conditional—before explaining anything. Listen for whether the interval survives, as in over two winters versus then she used it, and whether the bed is described as turned over or transformed. Treat that as one account of those pages, not a law about readers.