SEO Title Tag Variants That Fit the SERP
Generate title tags under 60 characters that carry the keyword and a real reason to click.
Generates multiple SEO title tag options for a page, each within the pixel-safe character range, front-loading the target keyword and adding a distinct click driver, with a note on which SERP position each option suits.
Ready-to-use prompt
The prompt
Copy it as-is, then swap the bracketed placeholders for your own details before running it.
Role: You are an SEO copywriter writing title tags.
Context:
- Page URL: {{page_url}}
- Page purpose: {{page_purpose}}
- Primary keyword: {{primary_keyword}}
- Secondary keywords: {{secondary_keywords}}
- Brand name: {{brand}}
- Competing titles on the SERP: {{competitor_titles}}
Task: Write 10 title tag options. For each give:
1. The title
2. Character count
3. The click driver used (number, year, outcome, contrast, question, authority)
4. How it differs from the competing titles
5. Best fit (top 3 position / positions 4-10 / new page)
Rules:
- 50-60 characters including the brand suffix if used.
- Primary keyword in the first 35 characters.
- No option may reuse another option's click driver.
- No clickbait the page does not deliver, and no invented numbers or dates.
- Include at least two options without the brand suffix.
- If the available driver categories cannot satisfy ten unique options, request clarification before drafting.
- Flag any keyword or suffix combination that makes the length constraints infeasible.
- Treat position labels as editorial fit, never as ranking predictions.
- Return titles as plain text, without wrapping quotation marks.
Finish with: your recommended title and a one-sentence reason.Estimated results
Editor's note
Why this prompt matters
Title tags are the highest-leverage SEO copy on any page and the most casually written. The usual output is ten near-identical variations of the keyword plus the brand name. This prompt fixes that by requiring a different click driver for each option — a number, an outcome, a contrast, an authority signal — and by making you paste the competing titles so differentiation is measured against the actual SERP rather than imagined. The position-fit column matters too: the title that wins from position eight is rarely the title that holds position one.
Anatomy
Prompt engineering breakdown
Role
Context
Goal
Constraints
Output format
What you'll get
Expected output
Worked example: a home composting guide
This illustrative page teaches beginners to set up a garden compost bin. Its supplied content includes seven setup steps, a materials checklist, advice on placement and layering, and troubleshooting for smells and pests. No brand suffix is required.
The primary keyword is compost bin setup. The example competitor set repeats broad beginner-guide and how-to patterns; it is not a live SERP sample. The opportunity is to make the practical help visible without promising faster decomposition or guaranteed results.
Resolve the driver conflict first
The original instructions request ten unique click drivers but name only six categories. Rather than quietly repeat categories or invent a year, the model flags the conflict. For this example, the editor explicitly approves the ten driver labels below before generation. Without that clarification, the model should pause and ask.
Candidate titles
Counts include spaces and punctuation. Every option places the complete primary keyword at the beginning.
| Title | Characters | Driver | Difference from example competitors | Best fit | |---|---:|---|---|---| | Compost Bin Setup: 7 Steps for Your First Garden Bin | 52 | Number | Makes the supplied process count explicit | Positions 4-10 | | Compost Bin Setup: Turn Kitchen Scraps Into Compost | 51 | Outcome | Leads with the practical end result | Top 3 position | | Compost Bin Setup: What to Add and What to Leave Out | 52 | Contrast | Highlights ingredient decisions | Positions 4-10 | | Compost Bin Setup: Why Does Your New Bin Smell Bad? | 51 | Question | Addresses a specific troubleshooting concern | Positions 4-10 | | Compost Bin Setup: A Checklist for Getting Started | 50 | Format | Advertises the usable checklist | New page | | Compost Bin Setup: A Beginner Guide for Home Gardens | 52 | Audience | Identifies both reader and setting | New page | | Compost Bin Setup: From Picking a Spot to First Fill | 52 | Scope | Defines where the instructions start and stop | Top 3 position | | Compost Bin Setup: Avoid Pests Before They Move In | 50 | Prevention | Emphasizes a supported practical concern | Positions 4-10 | | Compost Bin Setup: Layer Browns and Greens in Order | 51 | Sequence | Surfaces the layering instructions | Positions 4-10 | | Compost Bin Setup: Use the Materials You Have at Home | 53 | Resourcefulness | Highlights reusable household materials | New page |
Editorial decision
The smell question is distinctive, but it makes a supporting section sound like the page's main purpose. Save it for a dedicated troubleshooting page. Likewise, the layering option narrows the promise more than this broad guide needs.
Position labels are editorial hypotheses, not ranking predictions. A new page does not inherently need a different title structure. Check the actual query mix and displayed competing titles before treating these labels as priorities.
Recommended title: Compost Bin Setup: A Checklist for Getting Started — it distinguishes the page through a resource readers can actually use while preserving the broad setup intent.
Under the hood
Why this prompt works
Two constraints do the work. Forbidding reuse of a click driver stops the model producing ten paraphrases of the same idea, and requiring a differentiation note against real competing titles forces the output to earn its place on a specific SERP. The 50-60 character window with the keyword in the first 35 characters keeps titles from truncating and keeps the match visible where Google bolds it. The ban on invented numbers and dates prevents the most common quiet failure: a compelling title the page cannot deliver.
Model fit
Best AI models for this prompt
ChatGPT
Produces the widest range of genuinely distinct click drivers across ten options.
Claude
The most accurate character counting and the least willing to overpromise on the page's behalf.
Gemini
Best when you paste the live SERP titles and ask it to position against them directly.
When to use
- Before publishing, once the page content and purpose are final.
- After query-level review reveals strong impressions but weak click-through.
- When competing titles repeat a pattern your content can meaningfully challenge.
- During a title refresh where you can record the change and monitor results.
When not to use
- When an uneditable template controls the title.
- For paid ad headlines with different constraints.
- Before defining the page purpose and primary query.
- When you need current rankings or displayed titles; use live SERP observations and performance data.
- When technical indexing problems or title rewrites need human diagnosis first.
Get more from it
Pro tips
- 1
Paste actual competing titles alongside their URLs; ask which repeated pattern each candidate avoids.
- 2
Recount the final string in a spreadsheet with `LEN`, including spaces, punctuation, and any suffix.
- 3
Preview desktop and mobile rendering: 50-60 characters is a drafting constraint, not a pixel-safe guarantee.
- 4
Keep one separator maximum; remove stacked keywords rather than joining them with more pipes.
- 5
Reject unsupported promises. Supply relevant page excerpts so the model can verify each proposed driver.
- 6
Resolve the ten-options/six-drivers conflict before generation; approve extra categories or request fewer options.
- 7
For a second pass, identify the strongest candidates and specify exactly what failed: intent, differentiation, unsupported promise, or length.
Don't ship this
Common mistakes
✗ Ten variations of the same title.
Fix — Enforce the unique click-driver rule and reject duplicates.
✗ Burying the keyword after the brand name.
Fix — Keep the primary keyword inside the first 35 characters.
✗ Inventing a year or a number for freshness.
Fix — Only use figures the page actually contains.
People also ask
Frequently asked questions
Q.How long should a title tag be?
Aim for 50-60 characters. Beyond that Google truncates on desktop, and the truncation usually removes the part carrying your differentiation. The prompt returns a character count for every option.
Q.Why paste competing titles?
Because differentiation is only meaningful against the specific results you appear beside. Without them the model optimises against an imaginary SERP and hands you the same pattern everyone else is already using.
Q.Should the brand name be in the title?
Not always. On competitive informational queries the characters are better spent on the click driver. The prompt returns at least two brand-free options so you can compare both approaches.