How to write an SEO content brief
A content brief that actually helps a writer covers the gap to fill, the questions to answer and the proof to include, not just a keyword and a word count.
A working SEO content brief tells a writer the gap this page fills that competing pages do not, the exact questions to answer, the internal links and sources to use, and the intent and word band — everything needed to start writing immediately, without a clarifying question back to whoever assigned it.
It is the handoff point in a working SEO content strategy: strategy decides what to cover, the brief is what turns that decision into a page a writer can actually produce.
On this page
What does a content brief actually need to contain?
Seven things, in order: the target keyword and its intent, the specific gap this page fills, the questions it must answer, the internal links it must include, the sources to cite, the word band, and who reviews it before publishing. Anything beyond that is optional context; anything less than that leaves a real decision for the writer to guess at.
Notice what is deliberately not on that list. A brief does not need a full outline with every heading pre-written, because that turns the writer into a typist rather than someone making editorial judgment calls about how best to answer each question. It does not need brand voice guidelines repeated in full, because those belong in a separate, reusable style document rather than being copy-pasted into every brief. And it does not need a competitor's entire article summarised paragraph by paragraph — the gap analysis has already distilled the one thing that matters from that reading, and repeating the whole source just pads the document without adding a decision.
The test for whether something belongs in the brief is simple: does this remove a real choice the writer would otherwise have to make alone, or is it context they could look up themselves in thirty seconds. Keep the first kind. Cut the second kind ruthlessly, because a bloated brief gets skimmed, and a skimmed brief is functionally the same as no brief at all.
State the gap, not just the topic
"Write about keyword research tools" is a topic. "Cover Ahrefs, Semrush, Planner and Moz, and be honest about which is worth paying for at a small team's budget" is a gap.
List the exact questions to answer
Pull these from real searches, competitor headings and your own FAQ research — not from imagination.
Name the internal links required
The pillar, and any sibling posts the strategy depends on this page linking to, with the anchor text the link should use.
Attach two to five sources
Pre-verified links a writer can cite directly, so nobody is hunting for a source under deadline and citing whatever comes up first.
Give a word band tied to intent, not a fixed number
A definitional post and a comparison post need different lengths; a single house-wide word count produces padding on the short topics and thin coverage on the long ones.
Who should actually write the brief, and how long does it take?
Whoever did the keyword and competitor research for that topic — usually a strategist or senior editor, not the writer who will draft the page. This is a division of labour, not a hierarchy: the research step and the writing step use different skills, and asking one person to do both for every single page means the research gets rushed to leave time for the writing, or the other way around.
Expect thirty to sixty minutes per brief once you have a template and have already done the cluster-level keyword research the individual brief draws from. That estimate assumes the heavy lifting — the keyword map, the competitor read, the source list for the cluster as a whole — happened once, upstream, rather than being repeated from scratch for every page. A brief that takes three hours to write is usually a sign the upstream research was skipped, not that the brief itself is unusually demanding, and it is worth fixing the upstream gap rather than accepting the slower pace as normal.
Why does the "gap" line matter more than anything else in the brief?
Because it is the one thing a writer cannot reconstruct on their own without redoing the research the brief was supposed to save them from doing. Everything else in a brief — keyword, questions, links — a competent writer could approximate by skimming the top-ranking pages themselves. The gap requires someone to have actually read those pages, compared them and decided what none of them cover well, which is exactly the information gain work a good brief exists to hand off.
What does a weak brief look like next to a strong one?
| Weak brief | Strong brief |
|---|---|
| "Write 1,500 words on X." | "Cover these six specific questions; the current top results all miss Y." |
| No sources supplied | Two to five pre-verified sources with the exact claim each supports |
| No internal links specified | Named pillar and sibling links, with suggested anchor text |
| Word count as a hard target | Word band tied to intent, with a note that coverage beats padding |
The brief is where most of an editor's actual judgment lives. By the time a draft comes back, if the brief was vague, you are editing a guess — and no amount of line editing fixes a page that answered the wrong question well.
Does the brief change for AI-assisted drafting?
Less than people expect, and that is the point. A specific brief — gap, questions, sources, links — constrains an AI-assisted first draft the same way it constrains a human writer, which is why teams that already write good briefs tend to get usable AI-assisted drafts, and teams with vague briefs get generic ones whoever or whatever writes from them. If you are also checking output with AI content detectors, remember a detector score says nothing about whether the brief itself was any good — a specific brief followed carefully is what actually prevents generic output, not a scan afterward.
There is one addition worth making when a brief will feed an AI-assisted draft: name explicitly what the human editor is responsible for adding afterward. That might be a real client example, a number from your own data, or a quoted opinion from a named expert — something the model genuinely cannot produce, because it was not there. Putting that requirement in the brief itself, rather than leaving it as a vague hope for the editing pass, is what keeps an AI-assisted workflow from quietly drifting into publishing the model's first draft unchanged.
Does the brief format need to change for a second language?
Only slightly. Use one shared brief template across both languages so the process stays consistent, and add a field for language-specific keyword variants and any local examples the writer should reach for instead of directly translating the first language's choices. Our guide to bilingual content operations covers the full two-calendar setup; the brief itself is the one artifact that should look nearly identical in both languages, because its job — remove ambiguity for the writer — does not change with the language.
The one field worth adding specifically for a second language is a note on which examples, comparisons or references genuinely carry over and which need a local substitute. A pricing comparison, a regulatory reference, or a seasonal example that makes sense in one language's context can be confusing or simply wrong in the other, and a brief that flags this ahead of time saves a native-language editor from having to catch and explain it after a draft already exists.
How do you know the brief actually worked?
By checking what happened after the page published, not just whether the writer liked using it. Track which pages built from strong, gap-specific briefs went on to earn links, rank well or convert, against pages built from thinner ones — that comparison is the real feedback loop for improving your brief template, and it is the same data our guide to content ROI attribution shows you how to build.
Keep a simple record of which brief produced which page, even if it is only a link from the page's row in your content tracker back to the brief document. Six months in, when you are deciding whether to revise the template, that link is what lets you actually compare outcomes by brief quality instead of relying on memory or gut feeling about which pages "felt" like they had good direction going in.
Questions people also ask
How long should a content brief be?
Long enough to answer every question a writer would otherwise have to guess at, and no longer. One to two pages is typical: a paragraph of context, the gap to fill, the questions to answer, and the sources and internal links to use. A ten-page brief usually means the strategy work has not actually been done yet.
Who should write the content brief, the strategist or the writer?
The person who did the keyword and competitor research, because the brief's value is almost entirely in decisions that research produces. A writer building their own brief from scratch is redoing research that should have happened once, upstream, for the whole calendar.
Does a brief stop AI-assisted writing from being generic?
It helps more than almost anything else does, because a specific brief forces the specific decisions — the gap to fill, the examples to use, the person to quote — that a generic prompt skips. A vague brief and an AI tool produce the same generic result a vague prompt would.
Should a brief include the target word count?
A range, not a hard target. State the band the topic's intent calls for, but tell the writer explicitly that covering every required question well matters more than hitting a number, or you will get padding to reach it.
How do I know if a brief is good enough before handing it to a writer?
Read it as if you were the writer with no other context. If you could not start writing immediately without asking a clarifying question, the brief is missing something — usually the gap to fill or the specific proof points to include.
Sources
- How to Create an SEO Content Brief — Semrush (checked 23 September 2026)
- How to Create a Content Brief Template — Content Marketing Institute (checked 23 September 2026)
- Creating helpful, reliable, people-first content — Google Search Central (checked 23 September 2026)
Get briefs your writers can start from immediately
Book a free 30-minute consultation with Ibrahem and Jeremy. We look at your site while we talk, and you leave with a prioritised plan.
Who wrote this
Ibrahem HelalCo-founder
Leads strategy and delivery.
Works on: SEO, GEO
Reviewed before publishing by Jeremy Zava, Co-founder.
- SEO content strategy: the 2026 guideHow to plan, write, review, refresh and measure content that ranks in Google and gets cited by AI assistants, without tripping the spam policies.
- Do AI content detectors actually work?AI content detectors flag false positives and false negatives more often than their marketing pages suggest, according to independent testing.
- How to run bilingual content operationsA practical setup for running two content calendars in two languages without one becoming an afterthought translation of the other.
- How to attribute ROI to content marketingA practical way to connect content spend to revenue, using assisted conversions and cohort tracking instead of last-click credit alone.