Skip to content

AI Content Brief

Short answer

Content brief for one keyword: demand band, search intent, difficulty, AI-drafted titles and section openers to review, an outline, entities and a topic cluster.

Example:

Primary search term to analyse.

Example: best seo tools

Pick the geo target. Defaults to United States.

What It Does

Builds a writing brief for one keyword in one country. It estimates monthly search demand from several public signals, rates difficulty from the authority of the domains already in the top 10, classifies search intent and groups related phrases. An AI model then refines the title suggestions and writes a short summary and an opening line for each recommended subheading. The brief also lists entities to cover, a target word count and the domains that rank today.

Why It Matters

A writer who starts from a brief spends less time on research and is less likely to miss the subtopics readers expect. The brief puts the numbers next to the outline, so you can see how much demand a topic shows and how strong the current results are before anyone starts writing. The numbers are estimates from public signals: use them to compare topics, not as traffic forecasts.

How It Works

  1. Enter a keyword and choose a country

  2. We collect demand, results-page and topic signals in parallel; a signal that fails is skipped and shown in the signal coverage

  3. Combine them into a volume range with a confidence level, a difficulty score and an intent split

  4. Return the brief: titles, subheadings with opening lines, entities, a target word count and the ranking domains

Sample input + output

INPUT
keyword: "accessibility checker"
country: de
OUTPUT
Demand band        1K–10K (estimate)
Confidence         medium
Difficulty         41 · medium-easy
Intent             commercial 62% · informational 38%

Title suggestions
  1. Accessibility Checker: Test a Page Against WCAG 2.1
  2. How to Choose an Accessibility Checker

Target word count  1,600
Recommended H2 subheadings
  What an accessibility checker tests
  Automated checks vs manual testing
  Fixing the most common failures

Entities to cover  WCAG, EN 301 549, screen reader, alt text

Who Uses This

  • Content strategist

    Compare candidate topics by demand range and difficulty before adding them to the content calendar.

  • Editorial lead

    Give a writer a starting outline with title ideas, subheadings and entities to cover.

  • Freelance writer

    Agree the scope of an article with a client using the target word count and outline.

  • B2B SaaS founder

    Plan a product article in an afternoon, with the outline ready before you open the editor.

Frequently Asked Questions

What do I need to enter?

A keyword and a country. The language of the title suggestions follows the country: Greece and Cyprus give Greek; Germany, Austria and Switzerland German; France, Belgium and Luxembourg French; Spain Spanish; Italy Italian; every other country English.

How long does it take?

Usually several seconds, because many signals are collected at once. A repeat request for the same keyword and country is served from cache and returns faster.

What is in the result?

An order-of-magnitude demand band with a confidence level, a difficulty score and tier, the intent split, title suggestions, a short summary, recommended subheadings with opening lines, a target word count, entities to cover, topic clusters, key phrases and the domains ranking today.

How accurate is the demand band?

It is an order-of-magnitude estimate from public demand signals, not a count of searches. Confidence is medium when two or more signals had data, low when only one did and none when none did. Use it to compare topics, not to forecast traffic.

How is difficulty worked out?

From the authority of the domains in the top 10 results and whether the results page shows a knowledge panel. If no authority data is available, no difficulty score is given.

Which parts are written by AI?

The title suggestions start from fixed templates and are refined by an AI model; the summary and the opening lines are written by the model. The subheadings come from phrases that several signals agree on. Review everything before you publish.