A manuscript is not a folder of poems. It is an argument made by accumulation — one poem setting up the next, each one easing the reader a little deeper in. You can write twenty good poems and still have a shapeless pile; you can write fifteen and have a book. The difference is rarely in the writing. It is in the assembly: the ordering, the cutting, the small decisions about where a reader is allowed to breathe. What follows is the process worth handing to any poet preparing a first submission to an independent press — not a formula, but a set of working methods that make a collection hold together under an editor's eye.

Spread everything out before you decide anything

Start with inventory, not order. Put every finished poem on the floor, or in a spreadsheet if you prefer your chaos to have columns: title, first line, length in pages, and one word for its mood or register. Include the poems you've published in journals, the ones you've read at events, and the ones you've never shown anyone. Leave nothing out at this stage; deciding a poem is weak before you've seen it beside its neighbours is premature.

Then read the whole pile aloud, in two or three sittings. You are listening for tics rather than judging quality. Does the same image keep arriving — a window, a river, a particular grandmother's kitchen? Are three poems circling one breakup, all of them ending on the same resigned turn? Repetition inside a poem reads as obsession; across a manuscript it reads as padding. Reading aloud is also the fastest way to find the poems that only looked finished on the page.

Find the spine of the book

Every collection an editor takes seriously has a spine — a pressure the poems keep responding to. It might be thematic: a parent's decline, a move between countries, work in a particular trade. It might be formal: all short lyrics, or a sequence interrupted by prose poems. It might be tonal: a book that stays dry and wry rather than reaching for grandeur.

Try to write the spine in one sentence. Not a blurb, not a pitch — a private line to tape above your desk. These poems are about what a family refuses to say out loud. That sentence becomes a test. When you're unsure whether a poem belongs, hold it against the sentence and see whether it answers, complicates, or ignores it. Ignoring isn't automatic grounds for cutting, but a poem that ignores the spine three times over usually belongs in the next book.

What you're avoiding is the thesis-driven manuscript, where every poem announces the theme and the reader is never trusted to make a connection. A good collection leaves gaps.

Sequence by conversation, not chronology

The order in which poems were written is almost never the order in which they should be read. The date a poem was finished tells the reader nothing; the poem it sits beside tells them everything.

The three-position test

Take two poems you think belong together and read them in both directions. One order will feel like a turn; the other will feel like a jump. The turn is usually right, because it gives the reader a transition they can feel without noticing. When neither order works, they probably don't belong adjacent.

The opening move

Look at how each poem begins. A run of five poems that all open on a landscape description will feel monotonous even if each is excellent. Alternate the way you enter: a scene, a sentence of speech, a fact, a question, a fragment. The same applies to endings — don't let one poem close on light and the next open on light.

First and last

The first poem's job is to be welcoming and representative. It should not necessarily be your best poem; placing the strongest piece first sets a bar the next few poems have to climb down from. Many editors will tell you the best poem sits third or fourth, once the reader is committed. The final poem's job is to close a door: it should feel like the last thing said, not simply the last thing you happen to have finished.

Build sections deliberately

Collections of forty pages or more usually need two to four sections. Fewer than two and a longer book feels undifferentiated; more than four and the reader resets so often the momentum never builds. A section break is a breath and a change of register — a shift in place, in time, in address, in form.

Give each section its own centre of gravity. Part two might hold the longer, stranger poems; part three the sequence about work. Then check the section lengths. Long sections in the middle are where weak poems hide, and editors can spot a padded second act quickly. If a section can lose two poems without collapsing, look hard at which two.

You don't need labelled parts. A single short poem can serve as a hinge between larger groups, and often does the work more gracefully than a heading.

Cut the poems that don't pull weight

Assume the manuscript is currently ten to fifteen percent too long. Ask of each poem:

• If I remove it, does any reader notice?
• Does it say something another poem already says, only less sharply?
• Does it exist mainly because it was published, or because someone once praised it?
• Would I keep it if I had to defend it to an editor in one sentence?

Keep the answers in writing. The poem you love and nobody else does is real, and it is usually the first one that should go. Save it in a folder called next book; nothing is destroyed.

Prepare the submission package

Editors read for the poems and then look for reasons the file will be difficult. Remove those reasons.

1. Title page with your name, contact details, and the manuscript's page count.
2. Table of contents that matches the actual order, with page numbers.
3. An acknowledgements page listing journals and anthologies where poems first appeared, if any — with thanks and no false modesty.
4. Consistent formatting throughout: one serif font, one size, one document. One poem per page, and avoid starting a poem at the bottom of a page where only two lines fit.
5. Page numbers on every page, and a header or footer with your surname and the title.
6. A short cover letter: what the book is, its length, relevant publication history, and one sentence of biography. Three short paragraphs is plenty.

Then read the press's guidelines twice. Some independent publishers want a single file; others ask for blind submissions with no name in the document; some specify a maximum page count, and most differ on simultaneous submissions. Most presses land somewhere between twenty and thirty pages for a chapbook and forty-eight to ninety for a full collection, but verify for each individual publisher rather than trusting a general rule.

Before you press send

Put the manuscript away for two weeks. When you return, print it. Read the whole thing in one sitting, in the order it will be read, and mark every place your attention drifts. Those marks are the manuscript's real problems, and they are almost always structural rather than poetic.

Then hand it to one reader you trust — not someone who will praise it, someone who will tell you where they got bored. Fix the order, not the poems, unless a poem is genuinely unfinished. Check the contents page against the file one final time, save a clean copy, and send it. A manuscript does not need to be perfect to be accepted. It needs to feel like a book someone made a decision about.

Photo: betül nur akyürek / Pexels