Write Stronger Headlines Like The NYTs With This Free AI Prompt/Agent

Most blog posts live and die in the scroll. The headline is the split‑second decision point that determines whether your ideas get a serious look or vanish in the feed. Over 7.5 million blog posts/articles are published every day, which means weak headlines are buried before they ever have a chance to perform. This article will show you how to turn an AI agent into your personal New York Times style headline editor, built on nine proven headline tactics, and give you the exact prompt to use.

Why Your Headlines May Be Getting Ignored

Digital audiences skim. They do not owe us their attention. Most creators respond by either getting vague or getting loud. Both approaches fail. Poynter’s guidance on headline writing is clear. Specificity, emotional resonance, and strong verbs are what pull readers in, not clever puns or mystery for its own sake.

On LinkedIn, for example, the problem compounds. Your headline is competing with job changes, thought leadership carousels, and breaking industry news. If the value is not obvious within a second or two, your post is skipped. Headlines that read like internal notes rather than promises to the reader simply do not earn clicks.

The goal is not drama. The goal is clarity plus consequence. Your headline should answer a simple question. Why should a decision maker, recruiter, or peer stop scrolling right now and read this.

What This Free AI Headline Agent Actually Does

The prompt you are getting turns your AI-du jour into a specialized agent that behaves like a senior headline writer at The New York Times, with deep expertise in digital media, SEO, audience psychology, and conversion oriented editorial strategy. It has one job. To generate high performing headlines that balance clarity, credibility, and click worthiness without slipping into low quality clickbait.

The most popular AI platforms are built precisely for this kind of focused workflow. You can give a space custom instructions, attach reference files or links, and then reuse that specialized agent across multiple threads and projects. Instead of rewriting your system prompt every time you want stronger headlines, this agent sits ready, waiting for each new article, summary, or working title you feed it.

The result. You get headline output that feels like it came from a newsroom, not a generic AI template.

Inside The NYT Style Headline Prompt

At the heart of this agent is Chad Pollitt’s nine-tactic framework for sharpening headlines, originally developed from his analysis of New York Times and other premium publications. Each tactic is a different lens you can use to frame your story. The prompt instructs the agent to generate three headlines per tactic, for a total of twenty seven options off a single input. If you want less or more output, simply say so in the prompt.

The power of this structure is in forced creativity. Instead of settling on the first acceptable headline, you force the AI to look at your idea nine different ways, each rooted in real newsroom practices.

Quick Tour Of The Nine Tactics

Here is how each tactic plays out in practical terms.

1. Highlight the tension
2. Highlight the turning point
3. Note the scale of the issue
4. Address the reader directly
5. Call out telling statistics
6. Use two sentence structure
7. Surface voices from the story
8. Assign blame
9. Use specific and evocative language

The prompt instructs the agent to apply these nine tactics while maintaining clarity and credibility, which keeps your headlines sharp without straying into tabloid territory.

How To Use The Prompt Step By Step

Follow these steps to put the agent to work.

1. Set up the space
2. Choose your AI model
3. Gather your input
4. Run the agent
5. Evaluate and refine
6. Save your winners

Why This Beats “Write A Headline About X” As A Prompt

Most AI headline prompts are lazy. They ask models to “write a catchy headline” and call it a day. The output is often vague, fluffy, or indistinguishable from the rest of the feed.

This agent solves that problem in several ways.

• Contextual specialization
• Structured creativity
• Audience awareness baked in

You move from hoping you get something decent to systematically generating headline candidates that respect reader intelligence and still earn the click.

Practical Example: From Flat To Newsroom Ready

Take a generic LinkedIn headline.

“New AI tool improves LinkedIn engagement.”

Run that through the agent and you will see how the tactics transform it. For example.

• Tension tactic
• Turning point tactic
• Statistic tactic

None of these are guesswork. They use the conflict, turning points, and scale that serious publications rely on to frame stories. The prompt simply codifies that discipline into something you can reuse on every post.

Best Practices For Headline Testing

Once you have better headlines, treat your publishing like a laboratory, not a lottery. Poynter’s work on headlines shows that specificity, emotional cues, and strong verbs are more likely to keep readers engaged.

Use a simple testing routine.

1. Rotate tactics
2. Watch engagement patterns
3. Keep a headline log
4. Avoid overfitting

The prompt and space give you structured inputs. Your testing process turns those inputs into evidence, which is exactly how a newsroom refines its instincts over time.

How To Access And Use The Free Prompt/Agent

The agent is available below. I call it the "Headline Writer Pro" agent. Copy the system instructions, and either use it as a one-off prompt or clone it into your own agent for regular use.

You are a senior headline writer at The New York Times, with deep expertise in digital media, SEO, audience psychology, and conversion-driven editorial strategy.

Your job is to generate high-performing headlines that balance clarity, credibility, and click-worthiness—without sounding like spam or low-quality clickbait.

You will be given ONE of the following inputs:

- A rough summary of the article
- A full article
- A working headline
- A URL

If a URL is provided, infer the likely content based on structure and topic. If the input is thin, make reasonable assumptions but prioritize clarity over speculation.
---

### Your Objective

Generate headline options using the **9 headline tactics from Chad Pollitt’s framework https://www.linkedin.com/pulse/9-tactics-sharpen-your-headlines-chad-pollitt-zwvsc/ **, ensuring each headline clearly communicates a compelling value proposition.

For EACH of the 9 tactics:

- Produce exactly 3 distinct headline options (27 total)
- Each headline must be specific, useful, and audience-aware
- Avoid vague, generic phrasing (e.g., “tips,” “things,” “ways” without specificity)
- Avoid hypey or low-credibility language (e.g., “ultimate,” “insane,” “shocking”)
---

### Headline Requirements

Every headline must:
- Communicate a clear benefit or outcome
- Be understandable in under 2 seconds
- Match search intent where relevant (SEO-aware phrasing)
- Use specificity (numbers, audience, outcomes, or constraints)
- Be written for a skeptical, busy reader
---

### Output Format

Organize output in a table:

Columns:
- Tactic Name
- Headline
- Value Proposition (1 sentence explaining why someone would click)
- Specificity Score (0–5)
- Clarity Score (0–5)
- All Scores should be color coded as follows 0=🔴 1=🟠 2=🟡 3=🟣 4=🔵 5=🟢

Rules:
- Include 3 rows per tactic (27 rows total)
- Scores must be justified by the headline itself, not arbitrary
- Make sure to include a space between each headline in the output table
---

### Tactics to Apply

Use these interpretations of the 9 tactics:
1. Specificity and Numbers
2. Audience Targeting
3. Clear Benefit/Outcome
4. Curiosity Gap (without vagueness)
5. Problem-Solution Framing
6. Contrarian or Counterintuitive Angle
7. Urgency or Timeliness
8. Authority or Credibility Framing
9. Keyword-Optimized / Search Intent Match
---

### Selection Step (Critical)

After generating all headlines:
1. Select the TOP 3 headlines overall
2. Rank them #1–#3: should be color coded as follows in this section 3=🔴 2=🟡 1=🟢
3. Explain:
   - Why each one works
   - Where it would perform best (SEO, social, email, etc.)
   - What tradeoff it makes (e.g., clarity vs curiosity)
---

### Self-Critique Step

Then evaluate your own output:
- What is the weakest headline in the entire set?
- Why does it fail?
- Rewrite it to improve performance
---

### Input

[PASTE INPUT HERE]
        

Once inside.

• Bookmark the agent in your browser or respective AI platform so it becomes part of your publishing routine.
• Create a short checklist for yourself. Summary ready, agent open, headlines generated, best three selected, final one tested.
• Share the agent/prompt with your team if you collaborate on content. Shared headline discipline improves brand consistency.

Treat this as a permanent part of your content marketing stack, not a one off experiment. The more you use it, the sharper your sense of what works will become.

Bring NYT Discipline To Your Content Distribution Feeds

Great headlines are not accidents. They are the result of clear thinking about tension, stakes, audience, and language. These nine tactics capture how elite newsrooms frame stories on a daily basis. Many platforms today give you an easy way to encode that discipline into an AI agent that is always available when you sit down to publish.

Use the free prompt and agent to level up your next post. Then watch what happens when your content distribution feeds start reading less like a random stream of thoughts and more like a curated front page.

Bonus: This Agent Works For Social, Too

Article headlines can be used for much more than just posts. The above prompt/agent can also help you create short, punchy social media posts, paid media headlines for landing pages, email subject lines, ebooks, case studies, and many others, too.

In fact, one thing I like to do is use one headline for my article, another for when it's posted in social feeds, and another when it's promoted via email. It works great for promoting landing pages and/or product pages in paid media. These content assets don't usually have good titles.