Briefs, clustering, internal linking and on-page work — not mass-produced pages. Below are 7 copy-ready prompts. Fill in the [BRACKETS], copy, and paste into ChatGPT, Claude, Gemini or any capable assistant.
The honest version: AI is useful in SEO for everything except the page itself. Research, clustering, briefs, gap analysis, internal linking, schema — all good. Publishing at volume is the pattern search engines now act against by name.
The 7 prompts
Build a keyword cluster around one topic
Group keywords into pages instead of writing one article per keyword.
Build a topic cluster. CORE TOPIC: [THE SUBJECT] MY SITE: [WHAT IT DOES, WHO IT SERVES] MY AUTHORITY LEVEL: [new site / some rankings / established in this niche] KEYWORDS I HAVE: [PASTE YOUR LIST, or 'none - suggest them'] WHAT I SELL OR WANT PEOPLE TO DO: [CONVERSION GOAL] Produce: 1. THE CLUSTER MAP - group keywords by SEARCH INTENT, not by word similarity. Keywords that want the same answer belong on one page, even if they look different. Keywords that look similar but want different answers need separate pages. For each group: the page it becomes, the primary keyword, the secondary keywords it can rank for, and the intent (informational / commercial / transactional / navigational). 2. THE PILLAR PAGE - the one broad page everything links to, and what it must cover. 3. THE SUPPORTING PAGES - each with: working title, URL slug, primary keyword, intent, and the specific question it answers. 4. THE CANNIBALISATION CHECK - any two pages in my plan that would compete for the same query. Merge them and say so. This is the most common self-inflicted SEO problem. 5. PRIORITY ORDER - given my stated authority level, which pages to write first. For a new site, start with the lowest-competition, most specific pages, not the head term. Explain the reasoning. 6. THE COMMERCIAL PATH - which pages sit closest to my conversion goal, and how the informational pages should link toward them. 7. WHAT NOT TO WRITE - keywords in my list that are not worth a page: too competitive for my authority, wrong intent for my business, or better served as a section of an existing page. Rules: - Do not recommend one page per keyword variation. That is the scaled-content pattern Google penalises. - Be honest about competitiveness. A new site will not rank for a head term, and saying otherwise wastes months.
What you get: An intent-grouped cluster map with a cannibalisation check, a priority order matched to your authority, and a do-not-write list.
Tip: Grouping by intent rather than by keyword similarity is the whole technique. 'Best CRM' and 'CRM software' want the same page; 'what is a CRM' wants a different one.
Write a page that matches search intent
Build content that answers what the searcher actually wanted.
Plan a page to rank for this keyword. TARGET KEYWORD: [KEYWORD] WHAT CURRENTLY RANKS: [PASTE THE TOP 5 TITLES AND WHAT EACH IS, or describe them] MY SITE: [WHAT IT IS, AUTHORITY LEVEL] WHAT I CAN OFFER THAT THEY CANNOT: [YOUR ADVANTAGE - data, experience, a tool, access] MY CONVERSION GOAL: [IF ANY] Produce: 1. THE INTENT READING - from what currently ranks, what does the searcher actually want? The SERP is Google telling you what it thinks the intent is; match it or do not rank. Classify: are the results guides, listicles, tools, product pages, comparisons, forum threads? 2. THE FORMAT REQUIRED - if the top results are all tools and I write an article, I will not rank regardless of quality. State plainly what format this keyword demands. 3. WHAT ALL THE TOP RESULTS COVER - the table stakes. Omit any of these and the page looks incomplete. 4. WHAT NONE OF THEM COVER - the gap. This is where the page can win. 5. THE OUTLINE - sections in the order the searcher needs them. Lead with the answer; searchers arriving from Google want it immediately, not after 600 words of context. 6. THE ANSWER BLOCK - the 40-60 word direct answer to the query, to sit immediately under the H1. Written to stand alone if quoted. 7. TITLE AND META - three title options under 60 characters, and a meta description under 155 that describes the page honestly rather than teasing. 8. WHAT WOULD MAKE THIS BETTER THAN THE TOP RESULT - specifically, using my stated advantage. If I have no advantage over the current results, say so plainly - that is a reason to pick a different keyword rather than to write a page that will not rank. 9. INTERNAL LINKS - what should link to this page, and where it should link. Do not recommend writing a longer version of what already ranks. Length is not a differentiator.
What you get: An intent and format reading from the live SERP, the table stakes, the gap, and an honest check on whether you can beat what ranks.
Tip: Point 2 saves the most wasted effort. If every top result is a free tool, no article will outrank them no matter how good it is.
Audit a page that is not ranking
Diagnose why a page is stuck.
This page is not ranking. Diagnose it. PAGE: """ [PASTE THE CONTENT] """ URL: [URL STRUCTURE] TARGET KEYWORD: [KEYWORD] CURRENT POSITION: [RANK, or 'not in top 100'] IMPRESSIONS AND CLICKS: [SEARCH CONSOLE DATA, if you have it] HOW OLD: [AGE] SITE AUTHORITY: [new / moderate / strong] WHAT CURRENTLY RANKS: [TOP 3 RESULTS] Diagnose in this order - the earlier problems make the later ones irrelevant: 1. IS IT INDEXED - if not, nothing else matters. What to check: Search Console coverage, noindex tags, robots.txt, canonical pointing elsewhere. 2. IMPRESSIONS BUT NO CLICKS - this is a title and meta problem, not a content problem. The page ranks; nobody is choosing it. Fix the title. 3. NO IMPRESSIONS AT ALL - either not indexed, or ranking beyond page 5 for everything, which means an authority or relevance problem. 4. INTENT MISMATCH - compare my page's format to what ranks. This is the most common cause of a good page not ranking. State plainly if the format is wrong. 5. CANNIBALISATION - do I have another page targeting the same query? Ask me, and explain how to check. 6. CONTENT QUALITY AGAINST THE COMPETITION - not length. Specifically: does it answer the question faster and more completely, does it have anything the others do not, is there evidence of first-hand experience? 7. ON-PAGE BASICS - keyword in title, H1, first 100 words, URL. Heading structure. Internal links pointing to it. These matter, but they are not usually the reason a page fails. 8. AUTHORITY - given my stated site strength and what ranks, is this keyword realistically winnable? Be honest. Sometimes the answer is that the page is fine and the keyword is out of reach. 9. AGE - if the page is under three months old on a newer site, the honest answer may be that it needs more time. End with: THE SINGLE MOST LIKELY CAUSE, and the one change to make.
What you get: An ordered diagnosis from indexation to authority, with the single most likely cause named and one change recommended.
Tip: Point 2 is the highest-ROI finding in SEO. Impressions with no clicks means the page already ranks and a better title is a twenty-minute fix.
Write titles and meta descriptions that get clicked
Improve click-through rate on pages that already rank.
Write titles and meta descriptions. PAGE: [WHAT IT IS ABOUT] TARGET KEYWORD: [KEYWORD] CURRENT TITLE AND META: [PASTE, or 'none'] CURRENT POSITION AND CTR: [IF KNOWN] COMPETING TITLES IN THE SERP: [PASTE THE TOP 5] BRAND NAME: [NAME, and whether to include it] WHAT THE PAGE ACTUALLY DELIVERS: [BE HONEST] Produce: 1. TITLE OPTIONS - five, each under 60 characters (say the count). Each must: - Contain the keyword, preferably near the front - Say something the competing titles do not. If every result says 'Complete Guide', do not. - Be accurate. An overclaiming title raises CTR and destroys the ranking through pogo-sticking. Vary the approach across the five: the direct answer, the specific number or detail, the differentiator, the year or freshness signal (only if genuinely current), and the plain descriptive. 2. META DESCRIPTIONS - three, each under 155 characters. Each must: - Extend the title rather than repeat it - Say what the reader gets, specifically - Include the keyword naturally, since matched terms are bolded - End with a reason to click, not a sales pitch 3. THE SERP SIMULATION - show how each title and meta pair would appear as a search result, so I can see it in context. 4. AGAINST THE COMPETITION - given the competing titles I pasted, which of my options stands out, and why. If all the results look the same, the differentiated one wins by contrast alone. 5. THE HONEST CHECK - for each option, does the page deliver what the title promises? Mark any that overclaim. 6. RECOMMENDATION - one title, one meta, with reasoning. Rules: - Google rewrites meta descriptions often. Write them anyway; they are used most of the time for branded and long-tail queries. - No 'Ultimate', no 'Everything You Need to Know', no 'The Complete Guide' unless genuinely differentiating in this SERP - No clickbait. It works once.
What you get: Five titles and three metas with character counts, a SERP simulation, a differentiation check and overclaiming flagged.
Tip: Point 4 matters more than the copywriting. In a SERP where all ten results say 'Complete Guide', the one that says something specific wins on contrast alone.
Plan internal linking across a site
Use your own pages to lift the ones that matter.
Plan internal linking for my site. SITE STRUCTURE: [PASTE YOUR PAGE LIST OR SITEMAP] MY MOST IMPORTANT PAGES: [THE ONES THAT MAKE MONEY OR MATTER MOST] PAGES THAT GET TRAFFIC: [YOUR TOP PAGES BY TRAFFIC] PAGES THAT RANK 11-20: [IF KNOWN] SITE SIZE: [NUMBER OF PAGES] Produce: 1. THE HIERARCHY - how pages should relate: pillar, cluster, supporting. Flag any important page more than three clicks from the homepage. 2. ORPHAN AND NEAR-ORPHAN PAGES - pages with no internal links or only one. These are invisible to crawlers and to users. List them and say what should link to each. 3. THE AUTHORITY FLOW - my high-traffic pages have link equity to pass. Map: which traffic pages should link to which important pages. This is the highest-value internal linking work and almost nobody does it deliberately. 4. THE 11-20 OPPORTUNITY - pages ranking just off page one benefit most from additional internal links. For each, name three existing pages that could link to it naturally. 5. THE LINK PLAN - a table: From page | To page | Suggested anchor text | Why this link makes sense to a reader 6. ANCHOR TEXT RULES - descriptive, varied, keyword-relevant without being repetitive. Never the same anchor to the same page from every source. Give the anchor variations for each important target page. 7. LINKS TO REMOVE - excessive linking dilutes. Flag: sitewide footer links to individual articles, navigation bloat, and any page with an unreasonable number of internal links. 8. WHAT NOT TO LINK - pages that should not receive internal links: thin pages, near-duplicates, anything you would rather Google ignored. Rules: - Every link must make sense to a human reader. Links inserted purely for SEO read as spam and get ignored by both readers and Google. - Do not recommend a related-posts plugin. Automated linking produces irrelevant links.
What you get: An authority-flow map from traffic pages to money pages, orphan pages identified, and a link plan with varied anchors.
Tip: Point 3 is the underused lever. You already have pages with authority; pointing them deliberately at pages that need it is free and most sites never do it.
Decide what to do with underperforming content
Prune, merge, update or leave a library of old pages.
Help me decide what to do with these underperforming pages. PAGES: """ [PASTE: URL, TITLE, TARGET KEYWORD, TRAFFIC, IMPRESSIONS, POSITION, AGE, WORD COUNT] """ SITE: [WHAT IT IS ABOUT] TOTAL PAGES ON SITE: [NUMBER] WHY I AM DOING THIS: [traffic declining / site feels bloated / preparing for a redesign] For each page, recommend one of: KEEP AS IS / UPDATE / MERGE / REDIRECT / DELETE / NOINDEX — with reasoning. Decision rules to apply: - IMPRESSIONS BUT NO CLICKS → UPDATE the title and meta first. The page works; the listing does not. - RANKING 11-30 → UPDATE. This is the highest-return work available; a page on page two is close. - TWO PAGES TARGETING THE SAME INTENT → MERGE into the stronger one, redirect the weaker. Cannibalisation is costing you both. - NO IMPRESSIONS, THIN, NO LINKS, NO TRAFFIC → DELETE or NOINDEX. Say which and why. - NO TRAFFIC BUT HAS BACKLINKS → REDIRECT to a relevant page, never delete. Say which page. - SEASONAL OR CYCLICAL → KEEP. Check whether the low traffic is just the wrong time of year. - UNDER 3 MONTHS OLD → KEEP. Too early to judge. - SERVES A PURPOSE OTHER THAN SEARCH (support, legal, conversion) → KEEP regardless of traffic. Do not judge a pricing page by its organic traffic. Then: 1. THE ACTION TABLE - every page with its recommendation and reason. 2. THE MERGE PLAN - for each merge: which page survives, what content moves across, and the redirect. 3. THE REDIRECT MAP - old URL to new URL. Every deleted page with links or traffic needs one. 4. WHAT TO DO FIRST - ordered by expected impact. 5. THE RISK WARNING - what could go wrong. Deleting pages that were quietly earning long-tail traffic, or redirecting to irrelevant pages, both cost more than they save. 6. WHAT NOT TO TOUCH - pages I might be tempted to prune that should be left alone. Be conservative about deletion. Noindex is reversible; deletion is not.
What you get: A per-page action table with reasoning, a merge and redirect plan, and an explicit do-not-touch list.
Tip: The 'no traffic but has backlinks' rule saves real value. Deleting a page that quietly holds your best backlink throws away authority you cannot buy back.
Write content that AI answer engines will cite
Get quoted by AI Overviews, ChatGPT and Perplexity.
Optimise this content to be cited by AI answer engines. CONTENT: """ [PASTE] """ THE QUESTIONS IT SHOULD ANSWER: [LIST] MY EXPERTISE ON THIS: [WHY YOU CAN SPEAK TO IT] ANY ORIGINAL DATA OR EXPERIENCE: [WHAT YOU HAVE THAT OTHERS DO NOT] First, the honest framing: there is no special markup or file that gets you into AI answers. Google has stated there is no required schema and that files like llms.txt neither help nor harm. What gets cited is content that is indexed, crawlable, specific, and says something quotable that is not available everywhere else. Work from that. Produce: 1. THE ANSWER BLOCKS - for each question I listed, a 40-70 word standalone answer to sit directly under a heading phrased as that question. Each must make complete sense quoted with no surrounding context. This is the single highest-impact change. 2. THE QUOTABILITY AUDIT - quote every sentence in my content that could be lifted as a standalone claim. If there are few, the content is too hedged or too vague to cite. Say so. 3. SPECIFICITY UPGRADES - vague claims that should be specific. 'Many businesses find this helpful' cannot be cited; 'in our test of 40 accounts, 31 saw the issue within two weeks' can. Where I have the specifics, write the replacement; where I do not, mark [NEED SPECIFIC]. 4. THE ORIGINAL CONTRIBUTION - from what I told you, what does this content say that is not on twenty other pages? Answer engines cite sources that add something. If the answer is nothing, no optimisation fixes that; say so plainly. 5. STRUCTURE - descriptive question-phrased headings, one idea per section, definitions stated plainly, and the conclusion near the top rather than the bottom. 6. EVIDENCE AND DATING - claims with a date, a method, a sample size, or a named source are far more citable. Add where possible; flag where missing. 7. AUTHOR AND CREDIBILITY SIGNALS - visible author, credentials relevant to this topic, last-updated date. These are checked by both quality raters and retrieval systems. 8. WHAT NOT TO DO - do not create an llms.txt, do not stuff schema that does not describe visible content, do not create a page per question variant. Those are wasted effort or actively harmful.
What you get: Standalone answer blocks, a quotability audit, specificity upgrades, and an honest check on whether the content adds anything citable.
Tip: Point 1 is most of the value. A 50-word answer that makes sense with no surrounding context is what gets lifted; a well-argued 400-word section is not.
Where AI actually helps here
- Clustering a keyword export into topics and mapping them to pages
- Content briefs that say what a page must cover to be worth ranking
- Finding internal linking opportunities and writing anchor text that is not ‘click here’
Where it falls down
- Search volume, difficulty and SERP data. It does not have them and will produce plausible numbers anyway
- Judging intent from a keyword string without looking at what actually ranks
- Writing the page. That is where the volume trap is
The mistake almost everyone makes: Trusting invented keyword data
Ask a model for search volumes and it will give you a tidy table of numbers that came from nowhere. It has no access to search data. Get volume and difficulty from a tool that measures them; use AI for the clustering, the intent reading and the brief.
Free tool: Prompt Builder
Runs in your browser. No sign-up, nothing uploaded.
Questions people ask
Does AI content rank on Google?
Google’s guidance is that it rewards helpful content however it is produced, and that it acts on scaled content abuse — content generated at volume primarily to game rankings. The distinction is the pattern of production, not the tool.
What is the best AI use in SEO?
Content briefs and internal linking. Both are high-leverage, both are work nobody enjoys, and neither puts generated text in front of a reader.