Content

How to write a content brief that gets a usable first draft

A content brief works when the writer can start without asking you a question. Five decisions get you there, and most templates bury all five.

July 17, 20265 min read
fountain pen on spiral book
Photo by Aaron Burden

A content brief works when the writer can start the first paragraph without asking you a question. Most briefs never get there. They carry a keyword, a word count, and a deadline, then leave the real decisions to whoever opens the doc on Monday morning.

Five decisions do the work: the query, the intent, the reader, the shape of the page, and the claim the piece has to prove. Settle those five and the first draft comes back usable. Skip them and you rewrite the piece yourself, which is the job you were trying to hand off.

Name one query, one intent, and one reader

Pick one primary query. One. A brief carrying eleven keywords tells the writer nothing about what the page is for, so they average across all eleven and the page lands nowhere.

The query is the easy half. Intent is what decides the format. Someone typing "content brief template" wants a thing to copy. Someone typing "how to write a content brief" wants a method they can repeat. Same topic, two different pages. A writer who guesses wrong ships something that reads well and ranks nowhere.

So do the guessing for them. Open the first page of results, read the top three, and write down what they have in common: format, angle, how deep they go. Compress that into one sentence the writer can act on. "Method piece, step-by-step, aimed at someone who has already read the beginner posts." That sentence is worth more than the rest of the template. We go deeper on the mechanics in how to match your page to search intent.

Then settle the reader, in one sentence. "Marketers" is not a reader. "Head of growth at a 20-person SaaS, runs content with one freelancer, has read every beginner post already" is a reader. The second version tells the writer what to skip, what to assume, and which examples will land.

The test: can the writer picture one person? If the sentence describes a segment, it describes nobody. Cut it down until it names a job, a situation, and a level of knowledge.

Give the outline, and the point of every section

Headings alone are not an outline. "H2: Benefits" tells the writer where to type. It doesn't tell them what to say, so they fill the space with whatever fits, and you get four paragraphs that circle the topic without landing on it.

Under every heading, write one line naming the takeaway. Not the subject of the section, the point of it. "H2: Word count. Takeaway: derive it from what ranks, never from a house rule." Now the writer knows the argument and can spend their effort on the prose instead of guessing at your intent.

This is also where you catch a bad plan cheaply. If you can't write the takeaway line, the section has no reason to exist. Delete it in the brief, where it costs you nothing, rather than in the draft.

Turn word count and links into decisions you already made

Word count picked in a vacuum is noise. Derive it. Content Harmony's walkthrough of their brief process shows the move: for one query the top-ranking posts averaged just shy of 1,800 words, so the brief asked for 2,000 to 2,500. The number came from the results page, not from a house style rule. They also note that building a brief by hand runs one to two hours, which is the real reason most teams ship briefs that settle nothing.

Links get the same treatment. Name them in the brief: anchor text and URL, listed out. A writer told to "add internal links where relevant" will add three links to your homepage. A writer handed two specific URLs and the anchor text for each will build the cluster you're actually trying to build. While you're there, check that the page you're briefing doesn't overlap a page you already have. One strong page beats the thin set, which is the case we made in write one page that ranks instead of ten that don't.

Say what the piece has to prove

Every page worth publishing argues something. The brief should say what. "Content briefs fail because they list requirements instead of settling decisions" is a claim. A writer can serve it, or push back on it, or find a better one while reporting the piece. Any of those beats a draft assembled from headings.

Without a claim you get the shape of an article with nothing inside it. Every section true, every section fine, nothing the reader carries away. That's the piece nobody links to.

Cut the content brief down to what changes the draft

Most brief templates grow until nobody fills them in. Semrush's guide to content briefs lists thirteen components: primary keyword, secondary keywords, search intent, editorial direction, word count, target audience, brand voice, internal links, external links, competitor links, headings, technical details, and the call to action. Every one of those earns its place on some brief. Almost none earn a place on all of them.

Run the test on each field: if the writer would produce the same paragraph whether or not you filled this in, the field is decoration. Brand voice on a piece for a writer who has written twenty pieces for you? Decoration. Competitor links on a query where the results page is three forum threads? Decoration. Keep what changes the draft.

A brief that settles the query, the intent, the reader, the outline with takeaways, the links, and the claim fits on one screen. It takes twenty minutes once you have the results page open. And it saves the afternoon you'd otherwise spend rewriting a draft that was never going to work.

If you want help turning this into a content plan your team can actually run, that's what we do.

Let us find your next growth channel.

Tell us where you want to grow. We will tell you, honestly, whether search is the fastest way to get there.