Content brief: what to put in one, and how to check a draft against it

What is a content brief?

A content brief is the document that tells a writer what one piece of content must achieve and how it will be judged before they start writing. It sets the target query and search intent, the reader, the angle, the evidence and links the piece needs, and the criteria the draft is checked against. A good brief is short, because it records decisions rather than research.

Content briefs are one part of a content system: the keyword map decides what could get a brief, the editorial calendar decides what gets one this month, the brief decides what gets written, and the review decides what gets published. The whole system, from plan to retirement, is the subject of the guide to content operations. Where that system produces pages at scale, the same rules apply with higher stakes, which is the subject of the guide to programmatic SEO without thin content.

Why most content briefs do not work

Most briefs fail in one of two directions. Some are a keyword, a title and a word count, which leaves the writer to guess the reader, the angle and the standard. Others are a twenty-field template filled with competitor headings and term lists, which the writer skims and the editor never checks against. Both end in the same place: a draft that misses, and a review conducted on taste.

The published advice tends toward the second failure. Content Harmony’s guide, for example, lists more than a dozen elements and advises that it is better to go slightly overboard than to leave out necessary information. That is reasonable for completeness. The cost is that the decisions that matter, the angle and the standard, are buried among fields a writer could have worked out alone.

Failure	What the writer gets	What to do instead
Keyword and word count only	A topic, a length and no idea who it is for or what would make it good	Add the reader, the angle and the acceptance criteria
Competitor headings copied in	An outline of what already ranks, which produces a copy of what already ranks	Use the ranking pages to find gaps, and brief the gap
Term lists as instructions	Words to fit in, with no reason for each	List the questions the page must answer; the terms follow
No sources	Claims the writer has to find or invent	Name the evidence the page must cite, or the data you will supply
No acceptance criteria	A review that becomes a matter of taste	State the tests the draft has to pass before anyone reads it for style
Written after the outline	A brief that describes the draft rather than directing it	Write the brief first and get it approved before the outline

There is no good published evidence on how much briefs reduce revisions; the claims on vendor pages come without data. What can be measured is your own process, which is covered at the end of this guide.

What to put in a content brief

A content brief needs eight things, in this order. The first five are decisions that need the strategist’s context. The last three are the standard the draft will be judged against.

Part	What it says	Who decides
Target query and intent	One primary query, the variants that belong to the same page, and the search intent, checked against the keyword map	SEO or content lead
Reader and job	Who arrives on this page, what they already know, and what they should be able to do after reading	Content lead, with sales or product
Angle	What this page will say that the pages already ranking do not, in one or two sentences	Content lead and subject expert
Questions to answer	The questions the page must answer, in the order a reader would ask them	Content lead, from the results page and customer conversations
Evidence	The sources the page must cite, the data you will supply, and the claims it must not make	Subject expert
Links	The pillar to link up to, the siblings to link across to, the commercial page, and the pages that will link to this one	SEO lead
Format and length	The page type, the components it needs, and the scope that sets its length	Content lead
Acceptance criteria	The tests the draft must pass before review for style	Editor

Target query and intent

One primary query per brief, never two. Before the brief is written, check the query against your keyword map, the list of which URL owns which intent. If a page already owns it, the work is an update to that page, not a new brief. This is the check that prevents most keyword cannibalization, which usually starts as two briefs written months apart. State the intent explicitly; the Intent Classifier labels a keyword list by intent and suggests the page type for each.

Reader and job

Describe the reader in terms of what they know and what they are trying to do, not demographics. A sentence such as ‘a marketing manager at a 50-person software company who has been asked to start a blog and has never written a brief’ tells a writer what to explain and what to skip. Then state the job: what the reader should be able to do when they finish.

Angle

The angle is the part most templates leave out and the part that decides whether the page deserves to rank. Search the target query, read the top results, and write down what they all say and what none of them says. The angle is the second list. Google’s helpful content guidance asks whether content provides original information, reporting, research or analysis, and whether it provides substantial value compared with other pages in search results. A brief that cannot name the angle is not ready.

Questions to answer

List questions, not headings. Questions tell the writer what the reader needs; headings copied from competitors tell them what other writers chose. Take them from the results page, from the People also ask box, and above all from sales calls and support tickets, which contain questions no keyword tool shows.

Evidence

Name the sources the page must cite, supply any data of your own, and list the claims it must not make. This is where briefs most often fall short, and the gap gets filled with figures copied from other blogs. If a number matters to the argument, the brief says where it comes from.

Links

Specify the internal links as part of the brief, not as an afterthought in the edit: the pillar to link up to, two or three siblings, the commercial page where it fits, and the pages that should link to this one on publish day. The block to paste into a brief template is in the internal linking strategy guide. Google’s link best practices ask for anchor text that is descriptive, reasonably concise and relevant, so the brief can suggest anchors too.

Format and length

State the page type and the components it needs: a comparison table, a worked example, a calculator, a template. Then set the scope rather than a word count: the questions above, each answered fully, is the length. More on this below.

Acceptance criteria

The last section lists the tests the draft must pass before anyone reviews it for style. They are written as checks, so a writer can apply them before submitting. Examples are in the section on reviewing a draft.

How long should a content brief be?

A content brief should be as short as the decisions it records, which for most pages is one or two pages. Length in a brief usually comes from pasted research: competitor outlines, term lists, keyword exports. Put that in an appendix or a link, and keep the brief itself to what the writer must not get wrong.

A useful test is whether the writer could repeat the brief back in a minute: the query, the reader, the angle, the three or four most important questions, the sources and the bar for acceptance. If they could not, the brief is either missing a decision or hiding it.

Word count: why the brief should set scope, not length

Most brief templates include a target word count, usually the average length of the top-ranking pages. It is the wrong instruction. Google’s helpful content guidance asks whether you are writing to a particular word count because you have heard or read that Google has a preferred word count, and answers: no, it does not.

Length does correlate with results in surveys, which is why the habit persists. In Orbit Media’s 2025 survey of 808 content marketers, as reported by MarketingProfs, the average post was 1,333 words, only 9% published posts over 2,000 words, and 39% of those who did reported strong results, against 21% across all respondents. That is self-reported and a correlation: teams that write long posts may also research more, edit more and promote more. It is a reason to cover a topic fully, not a reason to stretch one.

Content brief templates for four types of page

The eight parts stay the same across content types, but what matters in each changes. These four templates cover most of what a B2B content team writes. Copy the one you need and delete what does not apply.

Guide

Comparison page

Glossary entry

Refresh of an existing page

The refresh template matters most for sites with an existing library: an update to a page that ranks is usually worth more than a new page

Refreshes deserve their own brief because they start from evidence a new page does not have: the queries the page already ranks for. Which pages to refresh, merge or retire is the subject of the content audit guide.

How to write a content brief, step by step

Writing a brief takes an hour or two for a standard guide, most of it in reading the results page and deciding the angle. The order below front-loads the checks that would stop the brief altogether.

1. 1CheckLook the query up in the keyword map. If a page already owns its intent, write a refresh brief for that page instead.
2. 2ReadRead the top results for the query. Note what they all cover and what none covers.
3. 3AskCollect the reader’s questions from the results page, sales calls and support tickets.
4. 4DecideWrite the angle in two sentences. If you cannot, stop and talk to a subject expert.
5. 5SourceList the evidence the page must cite and the data you will supply.
6. 6LinkFill in the links block: pillar, siblings, commercial page, inbound links.
7. 7ScopeSet the page type, components and what is out of scope.
8. 8AcceptWrite the acceptance criteria, then get the brief approved before the outline.

The content gap checker helps with the reading step: it compares a draft or a set of notes with the pages that rank and lists the terms and phrases they share that yours is missing. Use it to find what the ranking pages cover, then decide in the brief which of those gaps matter and which are padding.

Content briefs for AI-drafted content

When an AI model writes the first draft, the brief stops being guidance for a writer who will fill the gaps with their own knowledge and becomes the main source of everything specific in the page. A thin brief given to a person produces a thin draft that the writer improves. A thin brief given to a model produces fluent text built from the average of what it has read.

Google’s position is that how content is produced matters less than whether it helps people, with two limits a brief should respect. Its spam policies define scaled content abuse as many pages generated for the primary purpose of manipulating search rankings and not helping users, and list using generative AI tools to generate many pages without adding value for users as the first example. And its helpful content guidance asks whether the use of automation, including AI generation, is self-evident to visitors through disclosures or in other ways, and suggests disclosure where someone might reasonably ask how the content was created.

Brief for a writer

Strengths

• Can rely on the writer to research, interview and form a view
• Can leave the angle partly open for the writer to develop
• Evidence can be a list of sources to look at

Limits

• Still needs the query, reader, links and acceptance criteria

Shorter, because the writer supplies judgement.

Brief for an AI draft

Strengths

• Must supply the facts, figures and quotes the page uses, with sources
• Must state the angle and the claims to make, because the model will not find them
• Must list claims the model must not make, such as invented statistics or customer results

Limits

• Longer to write, and the review must check every fact against the brief’s sources

Longer, because the brief supplies the judgement.

The practical rule: everything in the draft that a reader could not have read elsewhere should trace back to something in the brief, an interview, your data or a named source. If a fact appears in the draft and not in the brief, the reviewer checks it before anything else. That rule applies with more force when briefs are templated across many pages, which is where the scaled content abuse policy starts to bite.

How to review a draft against the brief

A review against a brief has two passes. The first checks the acceptance criteria, which are pass or fail and can be done by anyone. The second, done only when the first passes, reviews the writing. Separating them is what stops a review turning into a rewrite.

Check	Passes when	If it fails
Answers the target query	The first paragraph answers the query directly, in words a reader would accept as an answer	Rewrite the opening before anything else
Delivers the angle	The point named in the brief’s angle is made clearly, and early enough to be found	Back to the writer with the angle restated
Answers every question	Each question in the brief is answered, in the order given or a better one the writer explains	List the missing questions
Evidence holds	Every figure and claim traces to a source in the brief, supplied data, or a source the writer adds and links	Source it or cut it
No banned claims	None of the claims the brief rules out appears, including invented statistics, ratings or testimonials	Remove the claim
Links in place	The pillar, sibling and commercial links are in the body with descriptive anchors	Add them before review
Scope kept	Out-of-scope topics are linked, not covered	Cut to a sentence and a link
House rules	Spelling, banned words and style guide checks pass	Fix before review

Write the acceptance criteria into every brief, and give the writer the same list to check before submitting. Most first drafts that fail a review fail the evidence or scope checks, both of which a writer can catch with the list in front of them.

Before publishing, the title and meta description are worth a separate check; the Meta Title Generator measures titles in pixels and flags the documented reasons Google rewrites them.

What content briefs cost, and what better ones save

A better brief costs more time up front and saves it in revision. Whether that trade pays depends on your own numbers: how many pieces you publish, how many revision rounds they take, and how long each round costs the writer and the editor together. The calculator puts those side by side.

Hypothetical defaults. Track your own revision rounds for a month before trusting the result.

Pieces published a monthRevision rounds per piece nowRevision rounds with better briefsHours per revision round, writer and editor togetherhoursExtra hours to write each better briefhoursBlended cost per hour$

Revision hours saved a month

24.0

Extra briefing hours a month

8.0

Net hours saved a month

16.0

Negative means better briefs cost more than they save at these inputs

Net value a month

$960

Revision rounds that must be saved per piece to break even

0.33

Revision rounds with better briefs are capped at the current number. The calculator counts time only; it does not value better pages, which is the larger effect and is measured separately.

The break-even line is the useful one. With these defaults, a better brief pays for itself if it saves a third of a revision round per piece. Most teams that count their rounds find that easy to clear, but count first: the point of the calculator is to replace an argument about briefs with a number from your own process.

How to tell whether your content briefs are working

Measure briefs at two points: when the draft comes in, and months after the page is published. The first tells you whether briefs are clear. The second tells you whether they choose the right pages and angles.

Measure	How to collect it	What it tells you
Revision rounds per piece	Count in your project tool, by brief author and content type	Whether briefs are clear enough for writers to hit
First-pass acceptance rate	Share of drafts that pass every acceptance check on submission	Whether the criteria are understood and achievable
Time from brief to publish	Dates in your project tool	Where pieces stall, and whether it is in briefing or review
Pages ranking for their target query	Search Console, each page’s target query, six and twelve months after publishing	Whether briefs pick winnable queries and angles
Cannibalisation incidents	New pages that trade places with an existing page for one query	Whether the keyword map check is being done

Review the numbers quarterly, by brief author and by content type. A brief author whose pieces take twice the revision rounds of everyone else’s usually needs a better template, not a talk. A content type whose pages rarely rank for their target query usually has an angle problem that starts in the brief. When attribution makes the business impact hard to read, the measurement approach in SEO measurement when attribution is already broken applies.

Content brief vs creative brief

A content brief and a creative brief do similar jobs for different work. A creative brief directs a campaign or an asset, such as an ad, a launch video or a brand refresh, and centres on the message, the audience’s current perception and the change the campaign should cause. A content brief directs one piece of editorial or search content and centres on the query, the reader’s question and the evidence.

Aspect	Content brief	Creative brief
Directs	One article, guide or page	A campaign or creative asset
Starts from	A search query or a reader question	A business objective and a message
Success looks like	The page answers the question, ranks and converts readers	The audience thinks or does something different
Key sections	Query, angle, evidence, links, acceptance criteria	Objective, audience insight, single message, mandatories
Usually written by	SEO or content lead	Strategist or account lead

Common content brief mistakes

Each mistake below shows up later as a revision round or a page that does not rank. Each comes with the question that catches it at briefing time.

• Two target queries in one brief. Check: is there exactly one primary query, and is it unowned in the keyword map?
• No angle. Check: can you say in two sentences what this page says that the ranking pages do not?
• Competitor outline as the brief. Check: are the sections questions a reader asks, or headings someone else chose?
• Word count instead of scope. Check: does the brief list what is in and out of scope?
• Evidence left to the writer. Check: are the key sources named, and the forbidden claims listed?
• Links added in the edit. Check: is the links block filled in before writing starts?
• No acceptance criteria. Check: could the writer test the draft against the brief before submitting?
• Same brief for a writer and for an AI draft. Check: if a model is drafting, does the brief supply every fact the page uses?

Decisions

One primary query, checked against the keyword map Search intent stated, with the page type that suits it Reader described by what they know and what they need to do Angle written in two sentences: what this page says that others do not

Content

Questions to answer, in reader order Sources to cite and data supplied, with forbidden claims listed Scope set, with out-of-scope topics and where they are linked Links block filled in: pillar, siblings, commercial page, inbound links

Standard

Acceptance criteria written as pass or fail checks Writer has the same checklist before submitting If AI drafts, every fact the page uses is in the brief or a named source

Ticks are saved in your browser.

Building briefing, review and publishing into one system, with templates and checks a team can run without the strategist in the room, is the work of a content systems engagement.