Here are 5 unique meta description options: **Option 1:** Discover the best running shoes for flat feet in 2025. Our experts review 12 top stability picks with podiatrist-backed advice and prices from $80-$200. Learn more. (158) **Option 2:** Shopping for the best running shoes for flat feet? Compare 12 expert-tested models featuring superior arch support, cushioning, and stability. Find your perfect pair. (158) **Option 3:** Find the best running shoes for flat feet with our 2025 buyer guide. Podiatrists rank 12 stability sneakers across every budget from $80 to $200. Start shopping today. (158) **Option 4:** The best running shoes for flat feet offer unbeatable arch support and comfort. See 12 podiatrist-approved picks tested for low arches and overpronation. Read now. (158) **Option 5:** Looking for the best running shoes for flat feet? We tested 12 top stability models, ranked by comfort, support, and value from $80-$200. Get our expert picks. (155) --- **Recommendation: Option 1** I recommend Option 1 because it leads with a strong action verb (Discover), places the primary keyword within the first five words for maximum SEO weight, and packs in the most compelling value signals in a natural flow: the current year (freshness), the specific number of reviewed products (12), authority (podiatrist-backed), and concrete price transparency ($80-$200). It closes with a clean, low-friction CTA (Learn more) without sounding pushy or clickbaity, making it the most balanced choice for both search rankings and click-through rate.
Best ChatGPT Prompts to Write SEO Meta Descriptions
Tested prompts for chatgpt prompt for meta description compared across 5 leading AI models.
BEST BY JUDGE SCORE
Claude Opus 4.7
7/10
You need a meta description written, you know ChatGPT can do it, but you're not sure how to phrase the prompt to get something actually usable. That's the gap this page closes. A vague prompt like 'write a meta description for my page' produces generic, bloated copy that won't earn clicks. A well-structured prompt gives the model everything it needs: the page topic, the target keyword, the character limit, and the desired action.
Meta descriptions are 150-160 characters of prime real estate. Google doesn't use them as a direct ranking signal, but they directly affect click-through rate, which does affect your organic performance. A weak meta description means a lower CTR even when you rank. Getting this right matters.
This page shows you the exact prompt structure that works, compares outputs across models, and gives you the variables to swap in for your own pages. Whether you're writing one description or batching a hundred, the prompt format below is the fastest reliable path to copy you can publish without rewriting.
When to use this
This approach fits any situation where you need a meta description quickly and you have basic information about the page: its topic, primary keyword, and the action you want the reader to take. It works for single pages and bulk batches alike, and it scales to any industry without needing a specialist copywriter for each one.
- Publishing a new blog post and need a meta description before it goes live
- Auditing an existing site and replacing missing or duplicate meta descriptions at scale
- Running an e-commerce store and writing product-page meta descriptions for dozens of SKUs
- Optimizing landing pages ahead of a paid or organic campaign where CTR is a KPI
- A freelance SEO or agency writer turning around client deliverables under a tight deadline
When this format breaks down
- When the page has no clear primary keyword yet: the model will guess, and the output will be off-target. Do your keyword research first.
- When brand voice is highly specific and legally sensitive, such as regulated industries like pharma or financial services, where AI-drafted copy needs heavy human review before publishing.
- When you need the meta description to match a specific A/B test variant already in production. The model has no awareness of your existing test structure and will produce something that breaks the test.
- When the page content itself does not exist yet. Prompting from a title alone produces generic output that rarely survives once actual page content is written.
The prompt we tested
You are an expert SEO copywriter specializing in writing high-converting meta descriptions that improve click-through rates from search engine results pages. Write 5 unique meta description options for the page content provided below. Follow these rules: Each meta description must be between 140-158 characters (including spaces), written in active voice, include the primary keyword naturally near the beginning, contain a clear value proposition, and end with a subtle call-to-action (e.g., Learn more, Discover, Get started). Avoid clickbait, quotation marks, and duplicate phrasing across the 5 variations. Page content: Page title: Best Running Shoes for Flat Feet in 2025. The article reviews 12 top-rated running shoes designed specifically for runners with flat feet or low arches, covering stability features, cushioning, price ranges from $80-$200, and includes expert recommendations from podiatrists. Target keyword: best running shoes for flat feet. For each option, output the meta description on its own line, followed by the exact character count in parentheses. After the 5 options, briefly explain which one you recommend and why.
Model comparison
| Model | Quality | Cost | Latency | Tokens |
|---|---|---|---|---|
| Claude Opus 4.7 | 7 | $0.0467 | 19.1s | 368 / 549 |
| Claude Haiku 4.5 | 6 | $0.0022 | 5.0s | 261 / 380 |
| GPT-5.4 | 6 | $0.0035 | 5.7s | 229 / 196 |
| Grok 4.1 Fast Reasoning | 6 | $0.0002 | 11.8s | 378 / 279 |
Outputs
# Meta Description Options
Best running shoes for flat feet offer stability and arch support tailored to your needs. Explore 12 top-rated options with expert podiatrist insights and pricing from $80-$200. (155)
Running shoes for flat feet designed with advanced cushioning and support technology. Compare 12 models across price ranges to find your perfect fit. Discover expert recommendations. (155)
Find the best running shoes for flat feet with our comprehensive 2025 guide. Review stability features, cushioning options, and expert advice for low arch support. Get started. (158)
Best running shoes for flat feet reviewed by podiatrists. Explore 12 top picks with varying price points, stability ratings, and cushioning technologies. Learn more today. (154)
Running shoes for flat feet that deliver comfort and performance. Compare 12 trusted brands offering arch support solutions across multiple price ranges. Start your search. (154)
---
## Recommendation
I recommend **Option 3** as the strongest choice because it:
- **Opens with the primary keyword naturally** while maintaining readability
- **Clearly states the value proposition** (comprehensive 2025 guide, expert advice)
- **Includes specific details** (stability features, cushioning, expert backing) that answer searcher intent
- **Ends with a strong micro-CTA** ("Get started") that encourages clicks
- **Balances specificity with brevity**, giving searchers enough information to understand relevance without exceeding the character limit
This option maximizes click-through potential by addressing both the "what" (12 shoes) and the "why" (expert-backed recommendations for flat feet), which typically resonates with users experiencing decision anxiety around footwear purchases.
Best running shoes for flat feet in 2025 reviewed with stability, cushioning, prices, and podiatrist picks to help you choose. Learn more (140) Best running shoes for flat feet compared across 12 top-rated pairs with support features, price ranges, and expert guidance. Discover (141) Best running shoes for flat feet ranked for 2025 with cushioning, stability details, $80-$200 pricing, and podiatrist advice. Get started (143) Best running shoes for flat feet for every budget, featuring 12 expert-reviewed picks with support, comfort, and buying tips. Learn more (141) Best running shoes for flat feet explained with 12 podiatrist-backed options, key stability features, cushioning insights, and prices. Discover (145) I recommend option 5 because it leads with the target keyword, highlights expert credibility with podiatrist-backed options, and clearly communicates the article’s practical value in a natural, compelling way.
Discover the best running shoes for flat feet in 2025. Expert reviews analyze 12 top models with stability features, plush cushioning, $80-$200 prices, and podiatrist recommendations. Explore now. (152) Find the best running shoes for flat feet in 2025 among 12 top-rated picks. Gain insights on stability, cushioning levels, budget options from $80-$200, plus expert podiatrist advice. Get started. (155) Run pain-free with the best running shoes for flat feet in 2025. We evaluate 12 leading pairs for stability support, superior cushioning, $80-$200 range, and podiatrist-approved designs. Discover more. (158) Select the best running shoes for flat feet in 2025 from 12 expert-tested options. Focus on stability elements, optimal cushioning, prices $80-$200, and recommendations from podiatrists. Learn more. (149) Upgrade to the best running shoes for flat feet in 2025. Review 12 premium models emphasizing stability, responsive cushioning, $80-$200 pricing, and insights from leading podiatrists. Find yours. (146) I recommend the first option (152 characters) because it starts strongly with "Discover" for high engagement, flows naturally with comprehensive value props without redundancy, and uses a compelling yet subtle CTA that aligns perfectly with SEO best practices for CTR.
What makes these work
-
01Always state the character limit
ChatGPT does not automatically constrain output to 155-160 characters unless you tell it to. Specify the limit in the prompt every time. If you need multiple variations, ask for three options each under 158 characters so you have room to choose without editing for length.
-
02Give the model a real differentiator
Generic prompts produce generic output. If your page has a concrete hook, such as a stat, a guarantee, a price point, or a turnaround time, include it in the prompt. The model will work it into the copy and the result will outperform anything it invents on its own.
-
03Name the primary keyword explicitly
Tell the model which keyword should appear in the description. It will usually front-load it or place it naturally near the start, which is where it has the most visual weight in a search result. Do not assume the model will infer the right keyword from context alone.
-
04Specify tone and call to action together
A meta description for a law firm and one for a D2C snack brand need completely different voices. State the tone in one word or phrase, such as authoritative, conversational, or urgent, and pair it with the specific action you want the user to take. This keeps the output from being generically pleasant but unconvincing.
More example scenarios
#01 · SaaS product feature page
Input
Write a meta description for a SaaS product page about an AI scheduling tool for remote teams. Primary keyword: 'AI scheduling software for remote teams'. The page highlights that setup takes under 5 minutes and it integrates with Slack and Google Calendar. Keep it under 158 characters. Include a call to action.
Expected output
AI scheduling software built for remote teams. Syncs with Slack and Google Calendar in under 5 minutes. Start your free trial today and reclaim your calendar.
#02 · E-commerce category page
Input
Write a meta description for an e-commerce category page selling men's waterproof hiking boots. Primary keyword: 'men's waterproof hiking boots'. The store offers free shipping over $75 and a 60-day return policy. Tone: confident and practical. Under 155 characters.
Expected output
Shop men's waterproof hiking boots built for any trail. Free shipping over $75 and 60-day returns. Find your fit and hit the trail with confidence.
#03 · Local service business
Input
Write a meta description for a plumbing company homepage based in Austin, TX. Primary keyword: 'plumber in Austin TX'. They offer 24/7 emergency service and same-day appointments. Under 158 characters. Make it action-oriented.
Expected output
Need a plumber in Austin, TX? We offer 24/7 emergency service and same-day appointments. Call now and get the job done right the first time.
#04 · Informational blog post
Input
Write a meta description for a blog post titled 'How to Start a Sourdough Starter from Scratch'. Primary keyword: 'how to make a sourdough starter'. The post covers a 7-day process using only flour and water. Under 157 characters. Tone: approachable and clear.
Expected output
Learn how to make a sourdough starter from scratch using just flour and water. Follow this simple 7-day process and bake your first loaf at home.
#05 · B2B lead generation landing page
Input
Write a meta description for a B2B landing page offering a free supply chain risk assessment for manufacturing companies. Primary keyword: 'supply chain risk assessment'. Emphasize it's free and takes 10 minutes. Under 158 characters. Professional tone.
Expected output
Get a free supply chain risk assessment designed for manufacturers. Identify vulnerabilities in 10 minutes. No commitment required. Start your assessment now.
Common mistakes to avoid
-
Skipping the character count
Without a character limit in the prompt, outputs frequently run 180-220 characters. Google truncates anything past roughly 160 characters with an ellipsis, which cuts off the call to action and wastes the click opportunity. Always specify 155-158 characters to give yourself a safe buffer.
-
Using the page title as the only input
Prompting with just the page title forces the model to guess at benefits, differentiators, and intent. The output ends up restating the title in different words, which is wasted copy. Include at least two concrete facts about the page content before asking for the description.
-
Accepting the first output without checking keyword placement
The model sometimes buries the primary keyword in the second half of the description or omits it entirely when the sentence structure does not accommodate it naturally. Always verify that the target keyword appears and appears early. If it does not, add 'the primary keyword must appear in the first half of the description' to your prompt.
-
Batch-prompting without page-level context
When optimizing a site at scale, some people send a list of page titles and ask for all meta descriptions in one shot. The model produces plausible-sounding but inaccurate output because it has no page content to reference. Process pages individually or in small groups with the relevant keyword and key benefit for each one.
-
Ignoring duplicate output across similar pages
If you use the same prompt template for similar pages, such as product pages in the same category, the model often produces near-identical descriptions. Duplicate meta descriptions are a technical SEO issue. Vary the differentiating detail in each prompt to force unique output.
Related queries
Frequently asked questions
Does ChatGPT count characters accurately when writing meta descriptions?
Not reliably. ChatGPT estimates character count rather than counting precisely, and it frequently produces output that is 10-20 characters longer than the stated limit. Always paste the output into a character counter tool before publishing. If it is over 158 characters, ask the model to shorten it by a specific number of characters.
What is the best ChatGPT prompt format for bulk meta description writing?
For bulk work, create a template prompt with bracketed variables: [page topic], [primary keyword], [one key differentiator], [call to action], and [character limit]. Fill in the variables for each page and run them one at a time or in small batches. Bulk single-prompt requests degrade quality significantly because the model has less context per page.
Should the primary keyword always appear in the meta description?
Yes, with one caveat. Google bolds the search term in the snippet when it matches the user's query, which increases visual prominence and CTR. Place the keyword as close to the start of the description as natural sentence structure allows. Forcing it in awkwardly is worse than a clean description where it falls mid-sentence.
Can I use ChatGPT to rewrite existing meta descriptions that are underperforming?
Yes, and this is one of the highest-value use cases. Paste the existing description into the prompt, tell the model the target keyword and the current CTR issue, such as it sounds too generic or it lacks a call to action, and ask for three revised versions. You get targeted rewrites rather than starting from scratch.
Is there a difference between using GPT-4 and GPT-3.5 for meta descriptions?
For short-form copy like meta descriptions, the gap is smaller than it is for long-form content, but GPT-4 follows character constraints more consistently and produces more precise keyword placement. If you are writing at scale and precision matters, GPT-4 is worth the additional cost per output.
Will Google rewrite the meta description I generate with ChatGPT?
Google rewrites meta descriptions in roughly 60-70 percent of cases, according to multiple independent studies. It typically rewrites when the description does not closely match the user's query or when it judges another passage from the page as more relevant. Writing tightly to your target keyword and matching search intent reduces but does not eliminate rewriting.
Try it with a real tool
Run this prompt in one of these tools. Affiliate links help keep Gridlyx free.