A template is a memory of the questions you skip when you are in a hurry. This one has two parts: a small file structure that holds a long-form project, and a one-page brief that sits inside it. Copy both, delete whatever your piece does not need, and fill in the brief before you open the draft.
- Keep six files per project: brief, research, outline, draft, decisions, archive.
- Fill the eight brief fields before opening the draft file.
- Write the central claim as a sentence somebody could argue with.
- Flag every research entry as verified or unverified as you record it.
- Cut to the parking lot and the archive rather than deleting outright.
What the template is for
Every long-form project asks roughly the same questions, and a rushed writer skips the uncomfortable ones. Who exactly is this for. What is the claim. What are you refusing to cover. A template does not make those questions easier to answer; it makes skipping them visible.
Keep it small enough to complete in twenty minutes. A template nobody finishes is worse than no template, because the empty fields become a standing reproach and the whole file eventually gets ignored. Delete any field you have never filled in twice.
The file structure
Six items sit at the top of a project: the brief, the research file, the outline, the draft, the decision log, and an archive. That is the entire structure. Each item has one job, and nothing is stored in two places at once.
Resist subfolders early. A project containing folders named misc and old has begun hiding things from you. When one item genuinely outgrows a single file, as research does on a book, split it by question rather than by source type. The archive is the exception and the pressure valve: anything cut or superseded goes there instead of being deleted, which makes removal feel reversible when revision demands it.
A template asks, on your calmest day, the questions you will avoid on your worst one.
The one-page brief, field by field
Working title, reader, the problem, the promise, the central claim, the desired response, length, and non-goals. Eight fields, and no more than that. The reader field should name somebody you could actually picture; busy professionals is not a reader, it is a line on a demographic slide.
The central claim is the field that most often exposes trouble. Write it as one sentence a reasonable person could disagree with. If the sentence is uncontroversial, you have a topic rather than an argument, and the draft will drift steadily toward summary.
Non-goals close the piece. Name three things you will not cover, including the tempting ones. Later, when a tangent arrives dressed as an important point, the line is already written and the decision has already been made by a calmer version of you.
The research file
One file, one entry per source, four things in every entry: what it says in your own words, the source and its date, why it matters to this piece, and whether you have verified it. The verification flag is the field that saves you under deadline.
Keep a short list of open questions at the top of the file and strike them out as they resolve. Anything still open when drafting begins becomes a marker in the text rather than a confident sentence you privately intend to check at some point.
The outline and the draft
The outline holds one full sentence per section, stating what the reader takes away. Underneath each sentence, list the research keys that support it. A section with a promise and no keys shows you exactly where research should go next.
The draft file carries the piece and nothing else. No notes to self in the body text, no pasted sources, no alternative paragraphs kept just in case. Cut material moves to the parking lot, and unresolved questions become markers or log entries.
The decision log and the parking lot
The decision log has three columns: date, decision, reason. It prevents the same argument being relitigated in week five, and it is the only part of the template that grows more valuable as a project gets older and memories get vaguer.
The parking lot holds good material this piece cannot carry. Keeping it visible reduces the urge to force a beloved paragraph into a section it does not serve, and it hands your next piece a genuine head start rather than a blank page.
Adapting the template
For anything under about 1,500 words, use the brief and the research file alone. For a report, add an approver and a distribution line to the brief. For a book, repeat the brief per chapter and keep one manuscript-level version sitting above them all.
Whatever else you cut, keep the reader, the claim, and the non-goals. Those three fields catch most of the trouble that would otherwise reach a draft, and together they take about six minutes to write on the day you start.
The template, ready to copy
How most projects begin: one file called draft, plus a folder of PDFs still carrying the names the downloads gave them.
PROJECT: [working title] 01-brief 02-research 03-outline 04-draft 05-decisions archive/ 01-brief Working title: Reader: one named person you could picture Their problem: Promise: what they can do or understand afterwards Central claim: one sentence someone could disagree with Desired response: Length and format: Non-goals: three subjects this piece will not cover Deadline and owner: 02-research Open questions, struck out as they close Entry: [key] / summary in your words / source and date / why it matters / verified yes or no 03-outline One sentence per section: what the reader takes away Research keys listed under each sentence 05-decisions Date | decision | reason Parking lot: material cut but worth keeping
Copy that block into a new project and complete the brief first. Ten minutes there is the cheapest hour you will ever save.
Common mistakes
| The mistake | Why it costs you, and what to do instead |
|---|---|
| Designing a template with forty fields | Every field looks useful while you are designing the form. In practice a long template gets abandoned around field twelve on the first busy day. Add a ninth field only after wishing for it twice. |
| Filling in the brief after drafting | Written last, the brief becomes a summary of whatever you happened to produce. Its value lies entirely in constraining the work beforehand. If you catch yourself back-filling, redraft the piece against it. |
| Skipping the non-goals | Non-goals feel negative and get left blank, which is why so many long-form pieces sprawl past their intended length. Naming excluded tangents is the only field that shortens a draft before a word of it exists. |
| Keeping research inside the draft file | Pasting sources into the manuscript saves a click and costs a fact check later. Keep entries in the research file with a verification flag and refer to them by key. The draft should hold publishable prose only. |
Your checklist
- Copy the six-file structure into a new project folder.
- Complete the brief before you create the draft file.
- Name one reader you could picture rather than a category of people.
- Phrase the central claim so that a reasonable person could argue against it.
- Write three non-goals, including the tangent you most want to take.
- Attach a verified or unverified flag to each research entry as you add it.
Key terms
- Brief
- A one-page file naming the reader, problem, promise, central claim, desired response, length, and non-goals of a piece, completed before drafting begins.
- Non-goal
- A subject the piece will deliberately not cover, written down in advance so a tempting tangent can be declined later without holding a fresh argument.
- Parking lot
- A section of the decision log holding material cut from the draft but worth keeping, so that removing a paragraph never feels like destroying it.
Setting a project up properly?
Frequently asked questions
Do I need all six files for every piece?
+
No. Below roughly 1,500 words the brief and the research file carry the work alone. Add the outline once a piece has more than four sections, and add the decision log once more than one person is able to change the direction of the work.
Should the brief be shared or private?
+
Share it whenever somebody else will approve, edit, or commission the piece. A brief read by one other person catches misunderstandings about the reader and the promise while they still cost minutes to fix rather than a full rewrite of a finished draft.
What goes in the archive rather than the bin?
+
Superseded drafts, cut sections, research you decided against, and any structure you abandoned along the way. Keep the archive flat and dated. Its whole purpose is to make revision feel reversible, which turns you into a far more willing cutter of your own work.
Can Strut AI fill in the template for me?
+
Strut AI can propose candidate fields from your notes: a possible claim, a list of open questions, three non-goals worth weighing. Treat that output as material to react to. The reader and the claim should be yours before any drafting starts.
How is the brief different from an outline?
+
The brief decides what a piece is for; the outline decides the order in which it happens. Writing the outline first is the most common cause of a well-structured article nobody needed. Settle the brief, then sequence what it commits you to.
What if the piece changes and the brief no longer matches?
+
Update the brief and record the change in the decision log. A brief that silently disagrees with the draft is worse than none at all, because reviewers will judge the piece against a promise you no longer intend to keep.