Best AI prompts for SEO: 10 copy-and-paste prompts

By Daniel Rowles, lead trainer at Target Internet and author of Digital Branding (Kogan Page). Last updated: November 2026.

The best AI prompts for SEO give the AI a clear role, real data from your own tools, numbered steps and a defined output, then tell it to flag anything it is guessing. Below are ten tried and tested copy-and-paste prompts covering keyword research, AI search visibility, content briefs, metadata, content refreshes and Search Console analysis.

Each prompt follows the same format: what it does, why it works, the prompt itself, and what to measure afterwards. Replace the text [in brackets] with your own details. This guide is part of our series of AI prompt guides for common digital marketing tasks.

How do you get better results from AI prompts for SEO?

AI tools are fast at structuring, drafting and spotting patterns, but on their own they do not know your search volumes, rankings or competitors. Unless your AI tool is connected to live data, anything it says about search demand is an estimate. The prompts in this guide work best when you paste in real data: a keyword export, Search Console queries, or the text of the pages that currently rank.

Use AI to make your content more useful, not to mass-produce pages. Google's spam policies treat using generative AI to create many pages without adding value for users as scaled content abuse, and the focus is on the lack of value, not the tool. Every prompt below keeps a human editor in charge of the final decision. For more on writing prompts in general, see our guide to perfecting your AI writing prompts.

AI prompts for SEO research and planning

1. Keyword clustering and intent mapping

What it does: Turns a raw keyword export into topic clusters, each with a search intent and a suggested page, so you can see where one page should target many related searches.

Why this works: It forces the AI to work only from your data, groups keywords by meaning rather than exact words, and flags clusters that would cannibalise existing pages.

Copy and paste prompt: "Act as a senior SEO strategist. Below is a keyword export with monthly search volume and current ranking URL where we have one. Group the keywords into topic clusters based on shared meaning and search intent, not shared words. For each cluster give: a cluster name, the primary keyword, the search intent (informational, commercial, transactional or navigational), the total volume, whether an existing URL already targets it, and a recommendation: new page, refresh existing page, or merge. Flag any cluster where two of our URLs compete for the same intent. Use only the data provided and do not estimate volumes. Our site: [domain]. Our audience: [audience]. Keyword data: [paste export]."

What to measure next: rankings and impressions for each cluster's primary keyword, and whether merged or refreshed pages gain visibility.

2. AI search prompt research

What it does: Builds a list of the conversational questions your audience is likely to ask AI assistants and AI search features, so you can check whether your content answers them.

Why this works: People phrase questions to AI tools as full sentences with context. Mapping those questions to your content shows gaps that keyword tools miss, and gives you a set of prompts to test your brand's visibility over time.

Copy and paste prompt: "Act as an AI search analyst. Our business is [business], selling [product or service] to [audience] in [market]. List 25 realistic questions this audience would ask an AI assistant at each stage of the buying journey: problem awareness, researching solutions, comparing options, and choosing a provider. Write them as people actually type or speak them, including context. For each question, name the page on our site that should answer it from this list: [paste URLs and titles], or mark it as a content gap. Group the gaps into no more than five content ideas."

What to measure next: run the questions in the main AI tools each month and record whether your brand is mentioned or cited. Our guide to measuring AI search visibility explains how to track this.

3. Search intent alignment and SEO refresh

What it does: Checks that a piece of content matches what people searching for its topic actually want, while keeping your brand's narrative.

Why this works: It builds an explicit intent hypothesis, maps missing entities and sub-topics, and produces two outlines, search-first and story-first, so you can choose the trade-off deliberately.

Copy and paste prompt: "Act as an SEO content strategist. Realign the following content to dominant search intent and modern on-page SEO best practice. Inputs: primary keyword or topic: [for example, B2B marketing automation]; secondary entities: [list]; target geography: [for example, UK and EU]; funnel stage: [awareness, consideration or decision]; KPI: [for example, free trial sign-ups]; brand tone: [for example, authoritative and approachable]; competitor angles to avoid: [if any]. Content to improve: [paste full text]. Tasks: 1) Define an intent hypothesis and your confidence in it. 2) Build a gap map of missing entities and sub-topics. 3) Propose Outline 1 (search-first) and Outline 2 (story-first) with the trade-offs. 4) Write snippet-ready answers and metadata. 5) Suggest internal linking anchors and structured data. 6) Provide two contrasting introductions."

What to measure next: keyword rankings and featured snippet capture, organic click-through rate, and clicks from internal links to your key pages.

4. SEO content brief

What it does: Turns your research into a brief a writer, or an AI, can work from, so new content is planned around search demand and information gain rather than written first and optimised later.

Why this works: It asks for what the current top results miss, which is where new content earns its place, and it builds in the evidence and first-hand experience that quality guidelines reward.

Copy and paste prompt: "Act as an SEO editor. Write a content brief for an article targeting [primary keyword] for [audience]. Here are the headings and key points from the current top five results: [paste]. Here are related questions from People Also Ask and our Search Console data: [paste]. Include: search intent; a working title under 60 characters; a 40 to 60 word direct answer for the opening; an outline of H2 and H3 headings phrased as questions where natural; what the current results miss that we could add, such as original data, examples or a clearer framework; where our own experience should appear; five FAQs; suggested internal links from this list: [paste URLs]; and a meta description of 150 to 160 characters."

What to measure next: time to first ranking, impressions in the first 90 days, and how often the page is cited in AI answers for its main questions.

If you want to build these skills with expert guidance, the Target Internet platform includes courses on SEO and AI, skills benchmarking, accredited certification, over 400 episodes of the Digital Marketing Podcast, and live masterclasses such as SEO, GEO and the future of search.

AI prompts for creating and optimising SEO content

5. Title tags and meta descriptions

What it does: Generates several title and description options for a page, each with a different angle, ready to test.

Why this works: It sets hard character limits, keeps the keyword near the front, and asks for distinct angles rather than small variations, which gives you real choices to test.

Copy and paste prompt: "Write five title tag and meta description pairs for this page. Primary keyword: [keyword]. Page summary: [paste the opening and headings]. Audience: [audience]. Rules: titles under 60 characters with the keyword near the start; descriptions of 150 to 160 characters that promise a specific outcome; British English; no clickbait or claims the page does not support. Make each pair use a different angle: direct answer, benefit, number or list, question, and comparison. Show the character count for each."

What to measure next: organic click-through rate in Search Console for the page before and after the change, compared over the same number of weeks.

6. Editorial refresh sprint

What it does: A structured edit that audits existing content for clarity, alignment with search intent and conversion impact.

Why this works: It uses a scoring rubric to identify weaknesses, produces a conservative and a transformative rewrite for testing, and builds in accessibility to maximise reach.

Copy and paste prompt: "You are a senior content editor for [industry or brand]. Your task is to diagnose and improve the following content to maximise clarity, usefulness, search alignment and conversion. Context: business goal: [for example, increase demo requests by 20%]; primary audience: [for example, UK mid-market HR leaders]; secondary audience: [optional]; product or service: [describe]; constraints: [legal, compliance, claims to avoid]; target keywords or themes: [list]; brand voice: [for example, practical, warm, expert; British English]. Content to improve: [paste full text or outline]. Process: 1) Audit the piece using a rubric scoring 0 to 5 on structure, clarity, E-E-A-T, search fit, completeness, brand voice, accessibility and conversion cues. 2) Identify the top 10 issues and opportunities and prioritise them with RICE. 3) Propose an improved outline. 4) Provide Version A (conservative edit) and Version B (transformative edit). 5) Create conversion assets: calls to action, metadata, FAQs and internal links. 6) Supply accessibility upgrades and a changelog."

What to measure next: form conversion rate, scroll depth and engaged time, keyword ranking improvements, and accessibility checks passed.

7. Brand voice, accessibility and persona rewrite

What it does: Rewrites content that is accurate but off-brand or hard to read, aligning it with your voice and making it more inclusive.

Why this works: It extracts a voice guide from your own samples, uses role-play critiques to surface objections, and produces two rewrites: one true to the brand and one tuned to a specific persona.

Copy and paste prompt: "You are Head of Brand and an accessibility reviewer. Rewrite content to match brand voice while improving inclusivity. Inputs: voice samples: [paste or list adjectives]; audience: [primary, secondary]; KPI: [for example, call bookings]; reading level: [for example, GCSE]; sensitivities or terms to avoid: [list]. Content to improve: [paste full text]. Process: 1) Extract a mini voice guide. 2) Run role-play critiques as a first-time reader and as a sceptical decision-maker. 3) Produce two rewrites: brand-true and persona-tuned. 4) Provide micro-improvements: calls to action, buttons and subject lines. 5) Add accessibility fixes: reading age, a plain-English summary and alt text."

What to measure next: readability score and time on page, click-through rate by audience segment, and consistency of brand language across content.

8. Competitor gap and differentiation builder

What it does: Finds weaknesses in competitor content and creates a distinctive stance so your page offers something the others do not.

Why this works: It creates side-by-side comparisons, labels gaps as table stakes, differentiators or thought leadership, and suggests original data or assets to strengthen your proof.

Copy and paste prompt: "Act as a category editor. Compare our content against competitors and build a differentiated, evidence-based upgrade plan. Inputs: audience and goal: [for example, UK SMB finance leaders; generate demo requests]; brand point of view: [for example, transparent pricing, strong support]; risk limits: [list]; competitor articles: [paste excerpts or outlines]; our article: [paste]. Tasks: 1) Create a comparison table with scores and short examples. 2) Identify 8 to 12 content gaps and classify each as table stakes, differentiator or thought leadership. 3) Draft a 100-word manifesto and unique stance. 4) Provide two headline and introduction options, one authority-led and one challenger. 5) Suggest one section only we can own and three original proof assets."

What to measure next: assisted conversions, backlinks earned from new data or assets, and engagement with your proof assets.

AI prompts for SEO measurement and distribution

9. Search Console performance analysis

What it does: Turns a Search Console export into a prioritised action list: pages close to page one, queries with high impressions but low clicks, and pages losing traffic.

Why this works: It gives the AI clear thresholds, asks for actions rather than observations, and keeps every recommendation tied to a specific page and query in your data.

Copy and paste prompt: "Act as an SEO analyst. Below is a Google Search Console export for the last three months with query, page, clicks, impressions, CTR and average position, plus the same data for the previous three months. Identify: 1) queries in positions 5 to 20 with at least [number] impressions, the striking-distance opportunities; 2) pages with high impressions but click-through rate well below others in similar positions, which need better titles and descriptions; 3) pages whose clicks fell by more than 20%, with the queries that dropped; 4) any two pages ranking for the same query. For each finding, recommend one specific action and rank all actions by likely impact. Use only the data provided. Data: [paste export]."

What to measure next: clicks and average position for the pages you act on, compared with the same period before the changes.

10. Content atomiser and refresh calendar

What it does: Turns one long-form asset into channel-specific pieces with testing built in, which brings more people to the original page and supports its visibility.

Why this works: It refactors the content for scanning and clarity, produces control and challenger versions of each piece, and sets up an eight-week calendar with tracked links.

Copy and paste prompt: "You are a content operations lead. Repurpose this long-form content into multi-channel derivatives with a refresh calendar. Inputs: audience and funnel stage: [for example, UK tech founders; consideration]; KPI: [for example, webinar sign-ups]; voice: [for example, confident and helpful]; channels: [LinkedIn, email, video, paid social]; key messages: [list]; constraints: [list]. Source content: [paste full text]. Tasks: 1) Refactor the article with an improved outline and executive summary. 2) Create derivatives for each channel. 3) Provide control and challenger versions with hypotheses. 4) Build an eight-week distribution calendar with cadence and owners. 5) Propose a UTM scheme and ready-to-paste links."

What to measure next: assisted conversions from derivatives, click-through rate by channel using your UTM tags, and video completion rates.

For more prompts in this series, see our guides to AI prompts for content planning and AI prompts to improve existing content. If you are ready to take a structured approach to building your digital marketing skills, explore Target Internet membership options and discover how the platform's interactive courses, benchmarking tools, podcast, accredited certification and live masterclasses can accelerate where you are heading next.

Frequently asked questions

What are the best AI prompts for SEO?

The best AI prompts for SEO give the AI a clear role, your own data, numbered steps and a defined output format, and tell it not to invent figures. The most useful cover keyword clustering, content briefs, titles and meta descriptions, content refreshes, competitor gaps and Search Console analysis.

Can ChatGPT do keyword research?

Partly. AI assistants are good at brainstorming topics, grouping keywords by meaning and suggesting search intent, but unless they are connected to an SEO data tool they do not know real search volumes or rankings. Use them to organise and interpret keyword data from a dedicated tool or Search Console, not to replace it.

How do AI prompts improve SEO?

AI prompts speed up the slow, structured parts of SEO: organising research, drafting briefs and metadata, spotting gaps against competitors and turning performance data into action lists. The improvement comes from acting on that output with human judgement, original expertise and real data, not from publishing AI text unchanged.

Does Google penalise AI-generated content?

Not because it is AI-generated. Google's spam policies target scaled content abuse: generating many pages, by any method, without adding value for users. Content that is helpful, accurate and reviewed by a knowledgeable person can rank whether or not AI helped produce it.

What is AI prompt research for SEO?

AI prompt research means identifying the questions people ask AI assistants and AI search features about your topic, then checking whether your content answers them and whether your brand is mentioned. It complements keyword research, because conversational questions are often longer and more specific than the keywords people type into search engines.

About the author

Daniel Rowles is lead trainer at Target Internet and the author of Digital Branding (Kogan Page). He co-hosts The Digital Marketing Podcast and delivers digital marketing and AI training for organisations around the world, including SEO, answer engine optimisation and AI for marketers. You can find out more on his Target Internet profile.

Bibliography

1. Google Search Central (2026). Spam policies for Google web search. https://developers.google.com/search/docs/essentials/spam-policies

2. Google Search Central (2023). Google Search's guidance about AI-generated content. https://developers.google.com/search/blog/2023/02/google-search-and-ai-content