Template · Practitioner
GEO Content Brief Template
A GEO content brief tells a writer which questions the page must answer, the direct answer to lead with, what original evidence to include, which primary sources to cite, who reviews it and how you'll check AI answers afterwards. It builds on a normal SEO brief. Google's guidance emphasises unique, helpful, people-first content (Google), so the brief's main job is to make sure the page has something worth citing before anyone writes.
Facts are from Google Search Central, accessed 30 Oct 2026. The template is our recommendation ("our view").
On this page
- Why use a brief at all?
- The template
- How do you fill in the brief well?
- What's a worked example?
- Which brief pitfalls should you watch for?
- How do you use the brief after publishing?
- How do you scale briefs across a team?
- How detailed should a brief be?
- How should writers and experts work together?
- Can you use the brief for refreshes?
- Checklist
- From MentionX data: why briefs should name the prompt intent
- Worked example: Ledgerline briefs a new guide
- How do you choose target prompts?
- What evidence should the brief require?
- How should the brief handle the brand?
- How long should the brief be?
- How do you use briefs with AI writing tools?
- What should the review step check?
- How do you record results in the brief?
- How do you store and share briefs?
- How do you keep briefs consistent across writers?
- Common questions
- Related
- Sources
Checked on 31 October 2026. "Our view" marks MentionX opinion. Ledgerline is a fictional company used for illustration.
Why use a brief at all?
Our view: most weak pages fail before drafting. Nobody decided what the page uniquely offers, which sources it relies on or who checks it. A brief forces those decisions early, when they're cheap. Google's AI optimization guide says non-commodity content, based on first-hand knowledge, matters most for generative AI search (Google). The brief is where you decide what that first-hand knowledge is.
The template
Copy this into your docs tool.
1. Basics
| Field | Fill in |
|---|---|
| Working title | |
| URL (new or existing) | |
| Page type | Guide / use case / comparison / docs / glossary |
| Owner | |
| Subject expert | |
| Reviewer and date | |
| Target publish date |
2. Audience and intent
- Who is reading? (role, situation, knowledge level)
- What are they trying to decide or do?
- What happens after they read? (next step)
3. Questions and prompts
| Type | Examples |
|---|---|
| Main question | |
| Follow-up questions (3–7) | |
| Prompts people might ask engines | |
| Search queries (from keyword tools or Search Console) |
Our view: Google describes its AI features fetching related "fan-out" queries (Google). List the follow-ups a reader would naturally ask next.
4. The direct answer
Write the 40–60 word answer to the main question here, before drafting. If you can't, the topic isn't clear enough yet.
5. What makes this page unique?
- First-hand experience or data we have:
- Expert perspective we can quote (name, role):
- Examples, screenshots or calculations we'll create:
- What competitors' pages miss:
6. Evidence and sources
| Claim | Primary source URL | Accessed |
|---|---|---|
Rules: primary sources only; no invented statistics; label fictional examples; mark interpretation as "our view".
7. Structure
- H2s as questions (list them)
- Tables needed
- Checklist or steps
- Images or video, with alt text plan (Google)
8. Product mentions
Where the product is relevant, and where it isn't. Disclose if comparing yourself to competitors.
9. Internal links
| Link to | Anchor |
|---|---|
10. Measurement
- Search Console metrics to watch (Google)
- Prompts to test, engines and frequency
- Business outcome (signups, demos)
- Review date
How do you fill in the brief well?
Our view: spend most time on sections 4 and 5. A clear answer and a genuine reason the page is better than what exists are what separate citable content from filler. Section 6 protects accuracy; insist that every factual claim has a source before drafting.
What's a worked example?
Fictional example: brief for "How to calculate holiday pay for part-time staff in Germany".
- Main question: how is holiday entitlement calculated for part-timers?
- Direct answer: drafted in 55 words, linking to the official German legal source.
- Unique: in-house payroll specialist's worked examples for three schedules.
- Sources: the relevant federal law text and a government ministry page.
- Measurement: monthly prompt tests in German and English.
Which brief pitfalls should you watch for?
- A keyword and word count with nothing else
- No named expert or reviewer
- Sources left for the writer to find later
- No plan for measuring results
- Target word counts; Google says it doesn't prefer particular word counts (Google)
How do you use the brief after publishing?
Our view: keep it with the page. At each review, check whether the answer is still correct, sources still load and the measurement plan shows progress. It becomes the page's record of why it exists.
How do you scale briefs across a team?
Our view: store the template in your docs tool, require it for every new page and major refresh, and have an editor approve sections 4–6 before drafting starts. Review a sample of published pages against their briefs each quarter to see where the process breaks down.
How detailed should a brief be?
Our view: detailed enough that a capable writer who has never spoken to you could draft a good page from it, and short enough to fit on one or two pages. If a brief runs much longer, it's often trying to cover several pages' worth of topics, which is a sign to split it. If it's only a keyword and a title, the writer will fill the gaps with generic material, which is exactly what makes a page easy for engines to ignore.
How should writers and experts work together?
Book a short interview with the subject expert before drafting, using the questions in section 3. Record it with permission, and pull quotes and examples from it into section 5. Our view: the expert then reviews the draft for accuracy only, not style, which keeps reviews quick and makes it easier to get busy experts to take part regularly.
Can you use the brief for refreshes?
Yes. For an existing page, fill in the brief using what the page currently says, then mark what's missing or outdated. The gaps become the refresh plan.
Checklist
- Questions and prompts listed
- Direct answer written first
- Unique contribution named
- Every claim has a primary source
- Expert and reviewer assigned
- Measurement plan set
From MentionX data: why briefs should name the prompt intent
Method. Read-only aggregate queries over MentionX scans from 22 August to 29 October 2026 (UTC): 2,217 engine runs (2,154 completed) across 18 anonymised projects and six engines.
Pages win mentions differently depending on the prompt they target. In our sample, brand mention rates varied sharply by intent:
| Intent | Brand mentioned | Share of voice |
|---|---|---|
| Brand-specific | 100% | 64.3 |
| Comparison | 91.9% | 44.6 |
| Recommendation | 29.2% | 26.8 |
| Best-of | 26.6% | 22.7 |
| Problem | 5.7% | 7.7 |
The cited page types also vary:
| Cited URL type | Share of 9,750 citations |
|---|---|
| Blog or article | 19.6% |
| Homepage | 10.0% |
| Review page | 2.7% |
| Docs | 2.6% |
| Pricing | 1.9% |
| Wikipedia | 1.6% |
Our view: a brief should say which intent the page targets. A page aimed at comparison prompts needs fair, sourced comparisons; one aimed at problem prompts needs a clear explanation where the brand is secondary. Mixing both in one page usually weakens it.
Limits. A convenience sample of mostly B2B software projects, run through engine APIs, which can behave differently from the consumer apps. Perplexity, Copilot and DeepSeek figures come from a single day (26 October 2026), so treat them as a snapshot, not a trend.
Worked example: Ledgerline briefs a new guide
Ledgerline is a fictional accounts-payable tool. Figures are illustrative.
Ledgerline's content lead writes a brief for a guide on three-way invoice matching. Target prompts: "what is three-way matching", "how to automate three-way matching" and "three-way matching vs two-way matching". Intent: problem and how-to.
The brief gives the direct answer in two sentences, lists five facts to include with sources, names a finance expert to review, sets a table comparing two-way and three-way matching, and notes that Ledgerline should be mentioned only in a short section on tools. It lists two competitor pages that are cited for these prompts and what they miss. After publishing, the writer checks the target prompts in three engines and records the results in the brief.
How do you choose target prompts?
Start from real buyer questions: sales calls, support tickets, search queries and community threads. Group similar questions and pick a primary prompt and two or three variants. Run them in the engines you track and note what's cited and what's missing. Those gaps are what the brief should ask the page to fill.
What evidence should the brief require?
List the facts the page must state and the source for each. Prefer primary sources: official documentation, laws, standards and your own published data. For each claim, note whether it needs a date. Facts without sources should not appear on the page, and the brief is the place to catch them.
How should the brief handle the brand?
Say where and how the brand should appear. On problem and how-to pages, keep it to a short, relevant section. On comparison pages, compare fairly and cite sources for every claim about rivals. Overly promotional pages are less likely to be cited and can reduce trust with buyers.
How long should the brief be?
One or two pages is usually enough. Include the purpose, target prompts, intent, audience, direct answer, key facts with sources, structure, expert reviewer, internal links and success measures. Longer briefs often repeat themselves and get skimmed.
How do you use briefs with AI writing tools?
If writers use AI tools for drafts, the brief becomes the input. Give the tool the direct answer, facts and sources, and ask it to follow the structure. Then have a human check every fact against the sources. Our view: AI drafts built from a good brief need far less fixing than drafts from a one-line prompt.
What should the review step check?
Check that the page answers the target prompts directly near the top, that every fact matches its source, that headings follow the planned structure and that internal links point to live pages. Ask the expert reviewer to sign off on accuracy.
How do you record results in the brief?
Add a results section at the bottom. After publishing, record the date, the engines checked, whether the page was cited and what the answers said. Repeat after a month and a quarter. This turns each brief into a small case study and helps you see which kinds of pages work.
How do you store and share briefs?
Keep briefs in a shared folder or your content tool, linked to the published page. Name them consistently. When someone refreshes the page later, they can see the original purpose and the results so far.
How do you keep briefs consistent across writers?
Use one template, with the same sections in the same order. Add one or two finished briefs as examples. Review new briefs for the first few weeks until the team is comfortable.
Common questions
Do I need a brief for every page? For important pages, yes. For small updates, a short checklist is enough.
Who should write the brief? Usually the content lead or GEO specialist, with input from an expert.
Should briefs include word counts? A rough range helps, but completeness matters more.
Related
- Rewriting an existing page for GEO
- Content refresh playbook for AI search
Sources
All accessed 30 Oct 2026.
Questions
What is a GEO content brief?
A one-page plan a writer uses before drafting, covering the questions the page answers, the direct answer, evidence and sources, structure, reviewer and how success will be measured.
How is it different from an SEO brief?
It keeps keyword research but adds the prompts people ask engines, the self-contained answer to lead with, the first-hand evidence to include and a plan for checking AI answers.
Who fills it in?
Usually an editor or strategist, with input from a subject expert. The writer should be able to draft from it without further meetings.