Content Brief Generator

What a content brief is, and who this is for

A content brief is the document you hand a writer before they write. Its job is to make the output aligned with what the page needs to do without requiring the writer to learn SEO, and without you rewriting the draft from scratch.

This is for anyone commissioning writing they will then have to edit: agency owners briefing freelancers, founders briefing a VA, and increasingly people briefing a language model rather than a person. The failure mode is identical in all three cases. A vague brief comes back as an article that reads fine and answers nothing, and you spend longer fixing it than you would have spent writing it.

The generator produces the brief. This page is about what to keep in it and what to throw away, because the standard content-brief template that circulates online contains at least two fields that Google has explicitly stated it does not use. Briefing to them wastes your writer's time in the best case and actively harms the page in the worst.

How to read the output: a worked example

Take the keyword make.com rate limit 429. The generator gives you something like this:

• Intent: informational, troubleshooting. Someone has hit an error and wants it to stop.
• Audience: technical, already using Make.com, mid-task and impatient.
• Questions to answer: what does 429 mean, why is Make throwing it, how do I add a delay, how do I retry, how do I stop it happening again.
• Suggested structure: what the error is, why the API returns it, the three fixes in order of effort, how to test the fix.
• Semantic terms: rate limit, retry, exponential backoff, error handler, requests per minute.
• Word count target: 2,450 words, based on the average of the top three ranking pages.
• Keyword usage: include the primary keyword 5 to 10 times.

The first five lines are useful. The last two are the problem.

The word count field

Google Search Central lists, among the warning signs that you are producing search-engine-first rather than people-first content, this question: Are you writing to a particular word count because you have heard or read that Google has a preferred word count? The answer given in the same document, in brackets, immediately after the question, is: No, we do not.

So the entire arithmetic chain behind the 2,450 figure, averaging three competitors and adding a margin, is built on a premise the search engine denies having. It is not a hard target and it is not a soft target. It is not a target.

Here is what happens when you brief it anyway. I have 117 published posts on this site, and the actual distribution of their lengths, measured rather than estimated, looks like this:

The shortest is 1,059 words and the longest is 3,751, a spread of more than three and a half times. I am not going to tell you that the long ones rank better, because I have not paired that distribution against per-post ranking data, and until someone does that properly the honest position is that this table shows you nothing about rankings at all. What it does show is what length looks like when it is decided by the subject rather than by a spreadsheet. The 1,059-word post is short because the question it answers is small. Briefing it to 2,450 words would have added 1,400 words of padding to a page whose whole value is that it gets to the fix quickly.

The useful replacement for the word count field is a question count. Instead of "2,450 words", write "answers these six questions, and stops". The writer then has a completion condition they can actually check, and the length falls out of it.

The keyword usage field

"Include the primary keyword 5 to 10 times" is the other field to delete, and it is the more dangerous of the two because it can push a page over a line that Google enforces rather than merely ignores.

The Google spam policies, last updated 15 May 2026, define keyword stuffing as filling a page with keywords in an attempt to manipulate rankings, and give as one example repeating the same words or phrases so often that it sounds unnatural. There is no threshold in that definition. The test is whether it sounds unnatural, which is a judgement a writer makes sentence by sentence and a brief cannot make in advance.

For a sense of what natural actually measures at, the post I published on that Make.com error contains the phrase "rate limit" 12 times in 1,280 words. That is 24 keyword words in 1,280, or a density of 1.88%. Nobody briefed that number. It is simply what happens when you write 1,280 words about rate limits, because the subject of the page is the phrase. If you find yourself needing a brief to tell the writer to mention the topic more, the real problem is that the page is not about the topic.

You can measure this on a draft rather than guess at it with the Keyword Density Checker. Use it as a smoke alarm, not a thermostat: a density of 4% or 5% means something has gone wrong, but there is no number you should be aiming for.

The method: brief from Google's own self-assessment list

There is a much better source for brief fields than a competitor average, and it is free and official. Google publishes a list of self-assessment questions creators can use to judge their own content, and almost every one of them converts directly into a line on a brief. That page was last updated on 10 December 2025.

The conversions worth making:

• Does the content provide original information, reporting, research, or analysis? becomes a brief field: what in this piece will not exist anywhere else? If the honest answer is nothing, the brief is not finished. This is the single highest-value line you can add, and it is the one no competitor-gap tool will generate for you, because by construction it cannot see things that have not been published.
• Does the content provide insightful analysis or interesting information that is beyond the obvious? becomes: name the non-obvious point. One sentence. If the writer cannot fill it in, they will produce a summary.
• If the content draws on other sources, does it avoid simply copying or rewriting those sources, and instead provide substantial additional value? becomes: which primary sources, and what does the piece add to them? Note the word primary. A brief that lists "top three ranking pages" as the reference material is briefing a rewrite of a rewrite.
• Does the content present information in a way that makes you want to trust it, such as clear sourcing, evidence of the expertise involved? becomes a required-citations count, and a named author. Google is unusually direct about bylines: it asks whether pages carry one where one might be expected, whether the byline leads to further information about the author, and states that it strongly encourages adding accurate authorship information.
• Does the content have any easily-verified factual errors? becomes a numbers-provenance line: every figure in the draft arrives with a source next to it, or it comes out.
• Is the content produced well, or does it appear sloppy or hastily produced? becomes the one thing most briefs leave implicit: who is doing the edit, and against what.

Google also frames the whole thing as three questions, Who, How and Why, and the Why is the one to put at the top of the brief rather than the bottom. If the honest answer to "why does this page exist" is "because the keyword had volume", the page will read that way no matter how good the structure section is.

One thing to be accurate about, because it gets overstated constantly: E-E-A-T is not a ranking factor. Google says so in that document, in those words, while also saying that a mix of signals which identify content with good E-E-A-T is useful, and that of the four aspects trust matters most. Anyone selling you an "E-E-A-T score" is selling you their own invention.

Which brief fields survive contact with the guidance

The pattern is worth stating plainly. Conventional briefs are strong on the fields that describe shape and silent on the fields that describe substance, because shape can be derived from pages that already exist and substance cannot. That is also precisely why so much briefed content ranks at parity and never above it.

The risk nobody puts in the brief

If you are generating briefs in volume and feeding them to a model, there is a policy you should read before you scale it rather than after.

Google's spam policies define scaled content abuse as generating many pages for the primary purpose of manipulating search rankings rather than helping users, typically by creating large amounts of unoriginal content that provides little to no value, no matter how it is created. The first example given is using generative AI tools to generate many pages without adding value for users. The second is stitching or combining content from different web pages without adding value.

Read those two examples next to a standard brief workflow: scrape the top three results, extract their headings, identify the gaps, generate an article that covers all of it. That is a description of stitching. The policy does not care that a human approved the brief.

What keeps you the right side of it is not volume, and it is not disclosure, though the guidance does suggest disclosing AI generation where a reader might reasonably wonder how a page was made. It is the originality field. If every brief you issue names something the piece will contain that does not exist elsewhere, and the finished pages actually contain it, the pages are not unoriginal, whatever produced the prose. If that field is blank on every brief, you are running the workflow the policy is describing.

Common mistakes

Briefing the shape and not the point. A brief with eleven H2s and no line saying what the piece argues will produce eleven sections of consensus.

Using competitor headings as the structure. The top three pages agreeing on a structure is evidence about what got published, not about what the reader needs. Copy it and the best available outcome is a fourth version of the same page.

Handing over a brief without the internal links. This is the part writers cannot do, because they do not know your site. Do the work: exact URLs, exact anchor text. Absolute URLs, so nothing breaks when the draft moves between tools.

Briefing keywords one at a time. Brief five closely related terms separately and you will get five pages competing for the same query and splitting whatever authority they earn. Cluster first. The Keyword Clustering Tool groups the terms that belong on one page, and the Keyword Research and Intent Analyzer tells you whether a term wants a blog post or a service page before you commission either.

No author. A page with no byline is unattributed content on the internet, which is the cheapest thing there is. If a freelancer wrote it, either credit them or have a named person on your side review it and take the byline honestly.

Rewriting the date instead of the content. Google lists changing the date of pages to make them seem fresh, when the content has not substantially changed, as a warning sign, and separately says that adding or removing content mainly to make a site seem fresh will not help rankings. It says so in brackets: no, it will not. Update the page or leave the date alone.

Treating the brief as the whole system. The brief is one document in a pipeline that also involves the keyword decision, the writer handoff, the edit, the internal linking pass and the publish. Briefs are usually the part people automate first because they are the most obviously templated, and the part that helps least, because the bottleneck is almost always the handoff.

Frequently Asked Questions

How long should a content brief be?

Long enough that the writer has no unanswered questions, which for most pieces is one page. If the brief is longer than the article, the brief is doing the writing.

What word count should I put in the brief?

None. Google states it does not have a preferred word count, and says writing to one because you believe otherwise is a warning sign that you are producing search-engine-first content. Brief the questions to answer instead and let the length follow. If you need a rough planning figure for scheduling or budgeting, use one, but keep it out of the writer's brief so it does not become a target to pad toward.

How many times should the keyword appear?

There is no correct number. Google defines keyword stuffing by whether repetition sounds unnatural rather than by a threshold, so no frequency is safe by arithmetic and none is required. Write the page about the subject and the phrase appears on its own. For reference, a post of mine on Make.com rate limits contains the phrase 12 times in 1,280 words, or 1.88% density, entirely without trying.

Can I give the brief straight to ChatGPT or Claude instead of a writer?

You can, and the brief will need to be better rather than worse. A model will follow a weak brief more literally than a person will, so a brief that says "cover these headings" gets you exactly those headings and nothing else. The fields that matter most in that scenario are the ones a model cannot invent: the original contribution, the primary sources, and the real numbers. Briefing a model is close enough to prompt engineering that the same principles apply, and there is a longer treatment of that in how to write a system prompt for AI agents.

Is AI-written content against Google's guidelines?

Not in itself. The spam policy targets scaled content abuse, defined as many pages generated primarily to manipulate rankings with little value to users, and it explicitly says this applies no matter how the content is created. So the question is not whether a model wrote it but whether the page is unoriginal and valueless, which is a property of the page rather than of the tool. Google also suggests disclosing AI generation where a reader might reasonably ask how the content was made.

Should the brief include competitor analysis?

It is worth having and worth demoting. Knowing what the top pages cover stops you omitting something obvious. Treating it as the specification caps you at parity with pages that already outrank you, which is a strange thing to aim for. Use it as a floor and put the original contribution above it.

Does a brief guarantee the content will rank?

No, and any tool that implies otherwise is overselling. A brief improves the odds that the draft is on-topic, complete and needs less editing. Ranking additionally depends on whether the site is trusted for the subject, whether the page is genuinely better than what is already there, and whether anyone links to it. What a brief reliably does is reduce your edit time, which is a real and measurable saving even in the months where rankings do not move.

What is the single most valuable field to add to a standard brief?

The original contribution: one sentence naming what this page will contain that does not exist anywhere else. It is the first item on Google's own content quality list, it is missing from nearly every brief template, and it is the only field that a competitor-derived workflow structurally cannot generate.

Where this tool stops

A brief generator produces a document. It does not move that document through a pipeline, and the pipeline is where content operations actually fail.

The pattern is consistent everywhere I have seen it. Briefs get written and sit in a folder. Drafts come back and wait for an editor who has other work. Approved pieces wait for someone to remember to publish them. The internal linking pass gets skipped because it happens after the exciting part is over. None of that is a briefing problem, and generating better briefs faster makes the queue longer rather than shorter.

The version that works is boring: a keyword row in a sheet or a table, a brief generated against it automatically, the writer notified with the brief attached, the returned draft routed to an editor with the checklist already applied, publication triggered on approval, and the internal links added from a lookup rather than from memory. That is an n8n or Make.com build with a handful of nodes, and it is the difference between a content process and a content habit.

That is the work I do: automation and AI pipelines that take a process someone is currently doing by hand and hold it together when they are busy. There is a walkthrough of the same architecture applied to distribution in how to automate social media posting, which is the same problem one step downstream.

If your briefs are good and nothing ships, that is the fixable part. You can see how I work on Upwork or Fiverr, or book a consultation and bring your current process, however held together it is.