Content briefs a writer can actually use

// In this article+

Content briefs are supposed to save a writer from guessing. Most of them still leave the decisions that matter most to guesswork. Semrush's guide to content briefs ↗︎ defines a content brief as a document that gives writers instructions and guidelines for creating content such as blog articles, white papers, or social posts. Clearscope's explainer on content briefs ↗︎ describes it as a detailed document that outlines the requirements, goals, and guidelines for a piece of content and serves as a roadmap for content creators. Both descriptions are accurate. Neither one locks the search intent to a fixed taxonomy, or pre-decides the internal links and anchor text. Neither names the entities and schema the page has to carry, or says what changes when a model drafts the piece instead of a person. A brief missing those four fields is a keyword list with a due date.

What is a content brief, and what makes one a writer can actually use?

A content brief is the single document a writer opens before drafting starts: the audience, the keyword, the format, and the outline. Semrush's content brief guide ↗︎ and Clearscope's content brief explainer ↗︎ both list keyword, audience, search intent, outline, word count, and internal or external links as the standard fields. That standard field list is where every content brief template ranking for this term stops.

The four fields most templates skip

Four fields turn a field list into a content brief template a writer can actually use, rather than one they still have to interpret on their own:

• Search intent labeled against a fixed taxonomy, not a vendor's paraphrase of one.
• Internal links with the anchor text already written, not a note to add internal links.
• The entities (specific, identifiable people, organizations, places, or concepts a search engine can resolve) the page has to name, and the one schema (structured markup that tells a search engine what kind of thing a page is) type it will carry.
• What changes, and what gets checked, when the draft is AI-drafted rather than human-drafted.

Brief vs. outline vs. style guide

A brief is not an outline, and it is not a style guide. An outline is the heading structure for one piece, and it can still change while a writer drafts. A style guide is standing guidance that applies to every piece a team publishes: tone, banned words, number formatting. A brief sits between the two, written once per piece, before the outline exists, and it hands the writer the intent, the links, the entities, and the schema. The outline can flex. A brief's locked fields should not.

How do you lock in the search intent before a writer starts?

Search intent is the job a searcher is trying to do when they type a query, not just the topic of the query. Most vendor guides describe intent with a four-way split: informational, navigational, transactional, and commercial. The original taxonomy is older and simpler. Andrei Broder's 2002 paper for the ACM SIGIR Forum ↗︎ argued that a web query is rarely just informational. Broder wrote that "it might be navigational (give me the url of the site I want to reach) or transactional (show me sites where I can perform a certain transaction, e.g. shop, download a file, or find a map)". Label a brief's intent field against Broder's three categories first, and treat commercial intent as a flavor of transactional.

Broder's informational, navigational and transactional queries

Most content marketing queries split cleanly across those three. "Content brief template" is informational: the searcher wants to understand the document. "Clearscope login" is navigational: the searcher wants one specific site. "Hire an SEO content writer" is transactional: the searcher is close to a decision. A brief that labels the wrong one sends the writer toward the wrong structure, an explainer where a comparison was needed instead.

Writing to the job the page does, not just the keyword

A two-line intent field is enough. Line one names the dominant intent, using Broder's three categories, and states the job the page does because of it: informational pieces explain, navigational pieces confirm identity, transactional pieces move a reader toward a decision. Line two names one secondary intent the piece should cover, usually inside the FAQ. This piece's own two-line field would read: dominant intent informational, explain what a usable content brief contains; secondary intent transactional, covered in the FAQ question about who should write one.

How does a brief decide the internal links and anchor text before the draft exists?

Internal linking is the set of links a site places between its own pages, and it is one of the few SEO decisions a brief can make in full before a writer opens a document. HubSpot's guide to topic clusters ↗︎ describes the model most sites now run. Every cluster page links to the pillar, and the pillar links to every important cluster page. That structure is what search marketers call topic clusters SEO: a pillar page owning a broad term. Narrower cluster pages link up to it and sideways to each other. A brief written inside that structure already knows which pillar a new piece links up to, before the writer is assigned.

Pillar-and-cluster linking, from HubSpot's own model

None of the templates ranking for "content brief" pre-decide anchor text. They say to include internal links and stop there. That leaves the writer to choose both the target page and the words that link to it. A brief written against the cluster map removes that choice and writes the anchor text into the document itself.

What Google says about anchor text

Google's own guidance on anchor text ↗︎ sets the bar the pre-written anchor text has to clear. It reads: "Good anchor text is descriptive, reasonably concise, and relevant to the page that it's on and to the page it links to". A brief that hands a writer "click here" has not done this part of the job.

A brief for this piece, tied to the content production cluster, would carry three pre-written links:

The writer never has to choose a target or invent anchor text. Both decisions were made before the brief reached them.

What entities and schema should the brief specify?

An entity is a specific, identifiable thing, a person, an organization, a place, or a concept, that a search engine or an AI model can resolve against its own knowledge graph rather than match as a string of text. Naming the entities a page has to mention is a fifth field most briefs never carry.

Naming the entities the page has to resolve

Google's own documentation on structured data ↗︎ states the purpose plainly: structured data helps Google "understand the content of the page". It also helps Google "gather information about the web and the world in general". A brief's entities field is a short, named list, not a paragraph. For a piece on content briefs, that list reads: Semrush, Clearscope, HubSpot, Andrei Broder, Schema.org, Nielsen Norman Group, Orbit Media. The writer's job is to mention each one by name, in context.

Picking one schema type before drafting, not after

Schema is the structured markup, written in a machine-readable format, that tells a search engine what kind of thing a page is. Schema.org's Article type ↗︎ carries its own properties, including articleBody, articleSection and wordCount. It also inherits CreativeWork properties such as author, datePublished, dateModified and headline. A brief picks the schema type before the writer starts, not after an editor notices the page has no markup. For most knowledge-base posts, that is Article. For a comparison piece or a template download, it might be a different type.

How should a brief set the 'helpful' bar and the boundaries for an AI-drafted piece?

Google's guidance on people-first content ↗︎ defines people-first content as content "that's created primarily for people, and not to manipulate search engine rankings". That sentence turns into three fields a brief can carry directly: who the page is for, why it exists beyond ranking, and what it should avoid. A brief that answers all three gives an AI-drafted piece the same boundary a human writer would otherwise infer.

Google's people-first questions, turned into brief fields

Three fields do the work. Audience: the intended reader, named specifically rather than as "marketers." Purpose: the specific decision or task the piece helps the reader make, beyond ranking. Boundary: what the piece should avoid, a comparison it should not turn into, a claim it cannot support, a tone it should not take. Write the three answers into the brief as short fields.

What to check before an AI-drafted piece goes out

Kevin Indig's piece on editing AI content ↗︎ frames the editing problem as "a workflow for the new age of content creation". Editing an AI draft is a distinct workflow, not a lighter version of editing a human one. A short checklist covers the gap the people-first fields do not close on their own:

• Every number and claim traces to a source in the brief, not one the model introduced.
• Every named entity from the brief's list appears, and no entity outside it got invented.
• The draft matches the pre-written intent field, not a different job the model chose.
• The pre-written internal links and anchor text survived the draft unchanged.

A brief this specific is what lets an AI-drafted first draft pass an edit instead of a rewrite.

How long should the piece be, and how should it be structured on the page?

Length and structure are the two fields most briefs treat as an afterthought. Eyetracking research and a running annual survey both give a brief something firmer to set that field against.

What the F-pattern means for paragraph order

Nielsen Norman Group's original eyetracking study ↗︎ found that readers scan pages in an F-shaped pattern: across the top, partway across a line lower down, then down the left side. Its guidance for writers follows directly. The first two paragraphs should state the most important information, and subheads, paragraphs, and bullet points should start with information-carrying words. A brief's structure field should say this outright: lead each section with the answer, not the setup.

Orbit Media's 2026 benchmark for length and time

Orbit Media's long-running annual blogging survey ↗︎ put the average blog post at 1,312 words, down from a peak of 1,427 words in 2023. The average time to write one fell to 3 hours, 20 minutes, down from a peak of 4:10 in 2022. Both numbers are falling for the same reason: faster drafting tools and tighter editing standards, not thinner content. Set a brief's length field at that 1,312-word median as a floor, not a ceiling.

What to do this week

1. Build a one-page content brief template with six fixed fields: search intent, entities to name, one schema type, three pre-written internal links with anchor text, and a length target.
2. Label intent using Broder's three-way taxonomy, informational, navigational, transactional, before a keyword gets assigned to the piece.
3. Pre-select three internal links for every brief: up to the pillar, sideways to a sibling cluster page, out to the relevant service page, anchor text written in rather than left to the writer.
4. Add a named entities list and a single schema type to the template, next to the keyword field rather than after it.
5. Run Google's three people-first questions, audience, purpose, and what to avoid, as a checklist against every draft, whether a person or a model wrote it.
6. Set the length and time target at Orbit Media's 1,312-word median as a floor, not a fixed ceiling.

FAQ

What's the difference between a content brief and a content outline?

An outline is the heading structure for one piece, the H2s and H3s and the order they run in, and it can change while a writer drafts. A brief is the set of decisions made before the outline exists: the audience, the intent, the entities, the schema, and the internal links with anchor text. The outline can flex. The brief's locked fields should not.

How long should a content brief be?

Long enough to answer the fields, short enough that a writer reads the whole thing. In practice that is one page: a target keyword, an intent field, an entities list, a schema type, three pre-written links, and a length target modeled on a current benchmark, such as Orbit Media's 1,312-word 2026 median.

Who should write the content brief, a strategist or the writer?

A strategist, or whoever owns the site's topic-cluster map, should write it, since the intent taxonomy, the entities list, and the internal links depend on decisions made above the level of one piece. A writer working from their own brief tends to pick the intent and the links that fit the draft already in their head, not the one the cluster needs.