15 Prompts for SEO Optimization: Meta Tags, Content, and Site Structure

15 Prompts for SEO Optimization: Meta Tags, Content, and Site Structure

Why AI Prompts Work for SEO

SEO is a systematic discipline: title tags, meta descriptions, heading hierarchy, internal links, structured data. AI language models can speed up the repetitive parts of this workflow if you give them clear, bounded instructions. The prompts below are based on official guidance from Google Search Central and the architecture of Schema.org. They do not replace a technical SEO audit — they help you produce and refine the textual assets that algorithms and users actually see.

According to Google, meta descriptions are not a direct ranking factor, but they can make your result stand out in search results and increase click-through rate. Similarly, heading tags help Google understand the structure of your page, and structured data makes your content eligible for rich results. This collection gives you 15 ready-to-paste prompts for three core areas: meta tags, content optimization, and site structure.

How to Use These Prompts

Each prompt uses placeholders like [KEYWORD] or [URL]. Replace the brackets with your own data. Then:

• Review the output for facts — AI can hallucinate.
• Use Google Search Console data to identify pages that need a CTR boost.
• Keep a spreadsheet to track the version of the tag you deploy.

Meta Tag Prompts (1–4)

1. Click-Worthy Title Tag Generator

Purpose: Generates multiple title tag options that respect Google display limits and your brand tone.

Prompt:

You are an SEO specialist. Create 5 title tags for the page [URL] targeting the keyword [KEYWORD].
Requirements:
- Include the keyword in the first 30 characters.
- Do not exceed 60 characters.
- Add a value proposition (e.g., Beginner-friendly, Updated for 2026).
- Do not use clickbait phrases like You won't believe.
- End with a recognizable modifier such as Guide, Checklist, Example, or the brand name.
For each title, explain why you chose it and estimate the expected click-through rate compared to a generic title.

Example input: [URL] = /blog/project-management-software, [KEYWORD] = AI project management

Example output:
1. AI Project Management Software: 2026 Buying Guide (58 chars) — includes keyword, year, guide format.
2. Best AI Project Management Tools (Free + Paid) (57 chars) — targets comparison intent.

Takeaway: Use the prompt on pages with less than 1% CTR in Google Search Console. Always preview the title to catch truncation.

2. Meta Description With a Clear CTA

Purpose: Produces compelling meta descriptions that include a call to action and a primary keyword without stuffing.

Prompt:

Write a meta description for [URL] that:
- Is between 140 and 160 characters.
- Starts with the active voice and mentions the main benefit for [TARGET AUDIENCE].
- Contains the keyword [KEYWORD] once.
- Ends with a clear CTA such as Learn more, Download the checklist, or Try it free.
- Avoids more than one exclamation mark.
Give me 3 options and mark the one that is most persuasive.

Example: Input: [KEYWORD] = Flesch reading ease, [TARGET AUDIENCE] = content writers.
Output: Improve your Flesch reading ease score in 10 minutes. Apply 5 simple editing rules that make your content more accessible. Start writing clearer today.

Takeaway: Google uses the whole snippet by default, so making the description relevant to the query is more important than exact character count.

3. Title and Description Compliance Checker

Purpose: Use a prompt to audit existing meta tags for length, duplicate titles, and missing tags.

Prompt:

Act as an SEO crawler. Below is a CSV list of pages with columns: URL, Title, Meta Description.
Check each line for:
- Title length > 60 characters or < 30 characters.
- Meta description length > 160 characters or empty.
- Duplicate titles across rows.
- Newline characters or unescaped quotes in tags.
Output a table with the page URL, issue type, and recommended fix. If the title may be truncated, propose a shorter title.

Example: You paste a Google Search Console export. The model returns a Markdown table. Use a dedicated crawler like Screaming Frog for the full site audit; this prompt helps you process the output quickly.

Takeaway: This is a time-saver for agencies, but it does not replace an SEO crawler.

4. A/B Test Proposal for CTR Improvement

Purpose: For pages receiving impressions but few clicks, this prompt suggests split-test hypotheses.

Prompt:

I have a page with an organic CTR of [CURRENT_CTR]% and [IMPRESSIONS] impressions per month.
Find the keyword [KEYWORD] in Google's SERP and look at the current result's position.
Propose 5 A/B test variants for the title tag and 5 for the meta description.
For each variant, specify:
- The exact new snippet.
- Why it should increase CTR (emotional hook, number, the word free).
- A possible negative effect (e.g., looks spammy).

Takeaway: Pair this with Google Optimize or another experimentation platform. Wait for a statistically significant number of clicks before making a decision.

Content Prompts (5–9)

5. Search Intent-Driven Outline Generator

Purpose: Builds a page outline that matches the dominant search intent to prevent keyword stuffing.

Prompt:

Determine the dominant search intent (informational, commercial, transactional, navigational) for the keyword [KEYWORD].
Based on that intent, create an H1-H6 outline for a new page with the title [TITLE].
Requirements:
- Include a brief definition in the intro.
- Add one section that compares or explains how to use [KEYWORD].
- Suggest a heading where a natural FAQ block fits.
- Write for an average reader with a reading level of 8th grade.
- The total outline should not exceed 10 headings.

Example: For how to create meta tags, the outline starts with What are meta tags?, Why they matter for SEO, etc.

Takeaway: Use this as a first draft, then cross-check with a manual SERP analysis.

6. Natural Language Variation Generator

Purpose: Creates LSI-style / synonym variations that improve lexical diversity and reduce keyword repetition.

Prompt:

List 15 natural language variations of the phrase [KEYWORD] that a real user would type in a Google search.
Group them into question form, action-oriented form, and long-tail modifier form (e.g. what is, best, for beginners).
Return them as a comma-separated list.

Example: For meta tags generator you might get: free meta tags generator, meta title generator, make meta description online — not exact synonyms, but related semantic concepts.

Takeaway: Use these variations in subheadings, image alt text, and body paragraphs. Do not force them.

7. Jargon and Readability Editor

Purpose: Rewrites text to meet a target reading level while preserving meaning.

Prompt:

Rewrite the following text for a [TARGET AUDIENCE] with a [READING LEVEL] reading level.
Break long sentences into two shorter ones.
Replace every acronym with the full phrase at least once.
Keep all factual claims unchanged.
Do not change the meaning of any sentence.
Text:
[PASTE TEXT]

Example: Paste an excerpt from a B2B page about machine learning and get a simplified version for a sales team.

Takeaway: Readability is an indirect SEO factor: easier text increases dwell time and reduces bounce rate.

8. FAQ Schema Content Trimmer

Purpose: Generates FAQ questions from a page and writes concise answers suitable for FAQPage markup.

Prompt:

Extract the 6 most likely user questions from the following page text. For each question:
- Answer in no more than 2-3 sentences.
- Use the exact keyword [KEYWORD] in at least 2 answers.
- Format the answer as plain text that can be inserted into FAQPage schema.
- Do not start every answer with the same phrase.
Page text:
[PASTE TEXT]

Takeaway: Make sure the FAQ content is actually visible on the page — Google may not show FAQ rich results if the answers are hidden.

9. Keyword Cluster Topic Map

Purpose: Turns a head term into 5–10 related subtopics for a topic cluster strategy.

Prompt:

You are an SEO strategist. The head term is [HEAD KEYWORD].
Generate a topic cluster by:
- Listing 10 long-tail keywords with monthly search potential (do not invent numbers).
- Identifying one pillar page URL and 5 cluster page URLs using slugs.
- Suggesting internal links from cluster pages to the pillar.
Return the result as a Markdown table with columns: sub-topic, suggested page slug, target keyword, and link candidate.

Takeaway: Use Google Search Console and Keyword Planner for real search volumes instead of trusting AI-generated numbers.

Site Structure Prompts (10–15)

10. Internal Link Plan Builder

Purpose: Creates a data-backed internal linking plan for a given page.

Prompt:

Given the current site structure (see list of URLs below), propose a set of internal links for the page [PAGE URL].
Rules:
- Use descriptive anchor text that contains the target keyword.
- Limit 3 links to the same page.
- Link from pages with high Page Authority to pages that need exposure.
- Do not link to pages that are noindexed.
List each suggestion in this format:
From page: [URL]
Anchor text: [ANCHOR]
To page: [URL]
Reason: [WHY]
URLs:
[List the URLs]

Takeaway: This prompt works best when you provide a raw crawl export. Without existing URLs, the output is generic.

11. Breadcrumb Schema Generator

Purpose: Produces JSON-LD for a breadcrumb trail.

Prompt:

Create JSON-LD BreadcrumbList schema for the following path:
Home > [CATEGORY] > [SUBPAGE].
Include each element with the position property and a valid URL from this list:
[URL HOME], [URL CATEGORY], [URL PAGE].
Output only the JSON-LD code, no explanation.

Example: Home > SEO Tools > Free Meta Tag Generator. The model outputs a valid BreadcrumbList object.

Takeaway: Validate the output in Google Rich Results Test — schema errors can be caused by trailing slashes.

12. URL Slug Cleaner

Purpose: Recommends SEO-friendly URL structures for existing pages.

Prompt:

Here is a list of URLs with unnecessary parameters or complex words. Suggest a cleaner URL for each one.
Requirements:
- Use lowercase letters, hyphens between words.
- Remove numbers, dates, stop words if the URL becomes shorter.
- Preserve the target keyword [KEYWORD].
Output a table with old URL, new URL, and 301 redirect recommendation.

Takeaway: Use a 301 redirect plan — do not simply change URLs without ensuring the old versions redirect.

13. Redirect Map Draft

Purpose: Helps during site migration to avoid 404s.

Prompt:

I am migrating a website. Below is an exported list of old URLs and the top new URL candidates. Map each old URL to a new URL.
- If you do not know the matching new page, put NO MATCH.
- Add for each mapping whether the redirect should be 301 or 302.
- Flag pages that should return 410 instead of 404.
Output Markdown table: Old URL

| New URL | Status | Reason.

Takeaway: Google advises 301 for permanent moves. Test everything on staging before going live.

14. Meta Robots Tag Auditor

Purpose: Creates correct noindex, nofollow, and canonical recommendations.

Prompt:

Analyze the following HTML snippets from the head of pages that are not ranking. For each page, tell me whether the robots meta tag is blocking Google from indexing the page, and whether the canonical URL is self-referencing.
Pages:
[PASTE HTML]
Suggest a concrete fix: remove noindex, update canonical, or keep as is.

Takeaway: Always verify with Google Search Console's URL Inspection tool. A noindex tag takes precedence over many other signals.

15. Lighthouse SEO Audit Interpreter

Purpose: Explain PageSpeed/SEO audit scores and turn them into a prioritised action list.

Prompt:

I ran Google Lighthouse for [URL]. Here are my scores:
Performance: [X], Accessibility: [Y], SEO: [Z].
Based solely on these scores, list 5 concrete actions that will improve the SEO score, in order of expected impact.
For each action, explain how to fix it in WordPress or general HTML.
Add a sentence that warns me about a possible side effect.

Takeaway: Use this on pages with an SEO score below 90. It is a starting point — do not rely on Lighthouse alone for technical SEO.

Case Study / Worked Example

Problem: A content marketer wanted to refresh 20 old blog posts that were not ranking above position 10 for their main keywords. The site had standard headings and long meta descriptions, so the marketer used prompt #1 and #5.

Solution: For each post, the marketer ran the title generator (prompt #1) and the outline generator (prompt #5) with the target keyword. They replaced old H2s with semantic variations from prompt #6, and used the FAQ trimmer (prompt #8) to add visible Q&A sections to six posts.

Result: Since we cannot present imaginary metrics, the actual result will depend on your niche and content quality. The main change was that meta descriptions became action-oriented — an adjustment that Google's documentation says can improve click-through rate. All pages were later re-submitted in Google Search Console.

Lesson: Prompts do not replace keyword research and a technical audit. They accelerate the copywriting part and help maintain consistency.

Conclusion

These 15 prompts cover the three pillars of on-page optimization: meta tags, content, and site structure. Use them as a scaffold, not a final product. Google's quality guidelines still require original expertise, accurate facts, and a good user experience. Start with prompt #3 or #5 to audit your existing pages, then apply the recommendations. If you combine AI output with a manual review in Google Search Console and the Rich Results Test, you will build a workflow that is both fast and reliable.

Sources

• Google Search Central: Meta tags — https://developers.google.com/search/docs/crawling-indexing/meta-tags
• Google Search Central: Redirects — https://developers.google.com/search/docs/crawling-indexing/301-redirects
• Schema.org: BreadcrumbList — https://schema.org/BreadcrumbList
• Google Rich Results Test — https://search.google.com/test/rich-results