Daily Money TreeTHE DIGITAL WORK NOTEBOOK
NEW WRITING + EARLY ARCHIVES

Writing & publishing

Write the content brief before the outline

Turn a vague subject into a reader task, an evidence plan and a reviewable piece of work.

1,158 words6 min readDaily Money Tree Editorial Desk

New editorial guide, published 17 October 2026. The historical sources linked below are preserved separately and are not represented as current product or service information.

A title can sound specific while the assignment behind it remains vague. An article about better project updates might be a beginner tutorial, a manager’s checklist, a software comparison or a sample message. Those are different jobs. Starting with an outline can hide that disagreement beneath a tidy set of headings. A short content brief makes the disagreement visible while changing direction is still inexpensive. The following method uses an invented project-update article as a working example. It is an editorial planning exercise, not a report about a real client or a promise of search traffic.

Name the moment, not an imaginary persona

Begin with the situation in which someone would need the page. In the example, a person must send a weekly update when a delivery date has slipped. They have ten minutes, several unresolved dependencies and no agreed wording for uncertainty. This description is more useful to the writer than a broad persona such as busy professional. It identifies the decision, the available information and the cost of ambiguity. It also suggests what the article must not assume: the reader cannot simply announce a new date if nobody has confirmed it.

Write one sentence describing what the reader should be able to do afterwards. Here it might be: produce a short update that separates completed work, risks, decisions and the next checkpoint. Then identify what the page will not resolve. It will not teach full project scheduling, negotiate a contract or recommend software. Exclusions keep a useful article from becoming a catalogue of loosely related advice. If the editor cannot explain why a proposed section helps that specific task, the section probably belongs elsewhere or needs a different purpose.

Turn the promise into observable output

A brief needs a visible result, not just an aspiration such as build confidence. Describe the output the article will help someone create. For the update example, that output is a message with four labelled parts and a clear request where a decision is needed. The article should contain a worked draft, an explanation of its choices and a way to adapt it. A reader who finishes without a usable message has not received the promised help, even if every sentence is pleasant and the introduction sounds reassuring.

Add a small acceptance test. Ask a reviewer to locate the current status, the unresolved issue, its owner and the next review date in the sample. If any item depends on inference, revise the example. These tests are editorial tools, not universal standards. Their value is that a writer and reviewer can apply them to the same text. Avoid acceptance criteria that secretly demand an outcome outside the article’s control, such as eliminating all stakeholder confusion. The page can make a message clearer; it cannot guarantee how every recipient responds.

Inventory evidence before arranging headings

Create three evidence columns: available, required and unavailable. Available evidence could include an approved status vocabulary, a public documentation page and an anonymised process description. Required evidence might include the meaning of a release stage. Unavailable evidence could be the reason for a particular delay. Do not fill that gap with a plausible story. Change the example or label the uncertainty. A brief that exposes missing evidence early can save an entire draft from being organised around a claim that nobody can substantiate.

Distinguish factual statements from proposed practice. The recommendation to put a decision request near the top is an editorial choice in this example, not a measured finding about every workplace. A claim that a method reduces missed deadlines by a stated percentage would require separate evidence with its population and method intact. Google’s people-first content guidance asks publishers to consider originality and usefulness; it does not establish a preferred article length. The length requirement for this edition is an editorial constraint, not a claim about how search ranking works.

Design the worked example and its boundaries

Invented examples are helpful when their status is obvious. Give the example only the details necessary to explain the decision. For instance, the draft can say that a dependency remains unconfirmed and that the next checkpoint is Tuesday. There is no need to invent a client’s revenue, a team’s impressive credentials or a dramatic failure. Use neutral placeholders for sensitive information. If a sample resembles a real project, remove identifying details and check that the remaining facts still support the lesson. A fictional label does not repair a misleading technical explanation.

Include a weak version and a better version for a specific reason. The weak message might say everything is progressing, despite an unresolved dependency. The revised message names the dependency and asks for a decision. Explain that change without treating the writer as careless or the recipient as incapable. The brief should also state the example’s limit: it assumes the update is permitted to disclose that dependency. Readers in different situations need to adapt the content, not copy a message whose assumptions do not apply to them.

Make review responsibilities explicit

Assign review by question rather than sending the draft to everyone for general thoughts. One reviewer checks whether the terminology matches the actual process. Another checks whether the sample makes sense without background knowledge. An editor checks scope, references and readability. A single person may do several jobs, but the questions should remain distinct. This makes conflicting comments easier to diagnose. A request for a longer explanation might reveal a missing assumption, while a request to remove that explanation might come from a different intended audience.

Record who can resolve each kind of disagreement. If nobody can confirm a factual assertion, the writer should remove or qualify it rather than treating silence as approval. Set a review window that the participants can actually meet, and record which version they reviewed. A comment on an earlier sample should not automatically block a later version that has changed the relevant premise. The brief does not need an elaborate workflow system. A short list of reviewers, their questions and the version identifier can provide enough structure for a small project.

Outline only after the assignment is stable

Now arrange the article around the reader’s task. The update example could move from gathering facts, to separating status from uncertainty, to writing the decision request, to reviewing the final message. Each heading should earn its place by supporting the promised output. Keep a parking area for worthwhile ideas that exceed scope. This avoids both losing useful thinking and inserting every idea into the current page. If the outline reveals that the task cannot be completed within the available evidence, return to the brief before drafting more prose.

Close the brief with a practical release checklist: the promised example exists, important assumptions are visible, factual claims have traceable support, references open, and someone unfamiliar with the project can follow the instructions. After publication, save questions that readers actually ask as possible revision evidence. Do not rewrite the brief retrospectively to make every result look planned. A good brief is a record of the assignment at a particular point, followed by documented changes. Its success is not that the first plan never changes, but that changes become understandable and reviewable.