How to Use AI to Write a Cold Email That Gets Replies
How to Use AI to Write a Cold Email That Gets Replies
Most cold emails fail because they are about the sender, not the recipient. AI tools do not fix bad strategy, but they make it significantly easier to write emails that are short, specific, and focused on the reader.
This guide covers the exact process for using AI to write cold emails that convert.
What Makes a Cold Email Work
Before you open an AI tool, understand what you are trying to achieve. A successful cold email does four things:
- Gets opened (subject line)
- Shows relevance in the first sentence (hook)
- Makes a specific, low-friction ask (call to action)
- Respects the reader's time (short length)
AI tools are useful for drafting and refining. They will not fix an email that lacks a genuine reason to reach out.
Step 1: Define Your Target Before You Prompt
The more specific your prompt, the better your output. Before opening an AI tool, define:
- Who you are sending to (role, industry, company size)
- Why you are reaching out (specific trigger or reason)
- What you want from the email (a call, a reply, a referral)
- One thing you know about this person or their company
A trigger is the most important element. Triggers include things like a recent funding round, a new product launch, a job posting, a social media post the recipient made, or a problem their industry commonly faces.
Step 2: Write the Prompt
Do not ask AI to "write a cold email." Ask it to write a specific email for a specific person with a specific goal.
Weak Prompt
"Write a cold email to a marketing director."
Strong Prompt
"Write a cold email from a freelance copywriter to a marketing director at a B2B SaaS company that recently launched a new product. The email should mention the product launch as a conversation opener, offer one specific way the copywriter could help with their product messaging, and end with a low-pressure ask for a 15-minute call. Keep the email under 120 words. Do not use a generic opener."
The strong prompt gives the AI a sender, a recipient, a trigger, a value proposition, a call to action, a word count, and a style instruction.
Step 3: Evaluate the Draft
Read the AI-generated draft and check for these problems:
Common Issues to Fix
Generic opener. If the email starts with "I hope this message finds you well" or "My name is X and I work at Y," rewrite the first line. The opener should reference something specific to the recipient.
Too much about you. Count the sentences that focus on the recipient versus the sender. If more than half are about you, revise.
Weak call to action. "Let me know if you are interested" is weak. "Would you have 15 minutes this week or next?" is specific and actionable.
Too long. Cold emails over 150 words have lower reply rates. If your draft is long, prompt the AI to cut it.
Step 4: Refine With Follow-Up Prompts
Use these follow-up prompts to improve the draft:
- "Rewrite the opening line to feel more like one person talking to another, not a sales email."
- "Cut this email to 100 words without losing the core message."
- "Rewrite the call to action so it is specific and low-pressure."
- "Make this sound less formal. The recipient is a startup founder."
Iterate until the email sounds like something a thoughtful person would actually write.
Step 5: Personalize Before Sending
AI can get you 80 percent there. The last 20 percent requires human judgment.
Before sending, add one genuinely personal detail. This should be something specific to the person you researched, not something the AI made up. Examples include referencing a talk they gave, a post they wrote, or a challenge their company publicly mentioned.
This one detail increases reply rates more than any other single change.
Writing Follow-Up Emails With AI
Most replies come from follow-ups, not first emails. If you have not heard back after five to seven days, send a short follow-up.
Prompt for a Follow-Up
"Write a three-sentence follow-up to this cold email. Acknowledge that the recipient is busy. Add one new piece of value or context that was not in the original. End with the same call to action."
Do not apologize for following up and do not ask if they received your email. Keep it brief and add value.
Best AI Tools for Cold Email Writing
ChatGPT
The most flexible option. Works well for custom prompts and iterating on drafts. Handles tone adjustments well.
Claude
Produces cleaner, more natural-sounding copy. Particularly good at removing the sales-y tone that makes cold emails easy to ignore.
Lavender
An AI tool built specifically for cold email. It scores your email in real time and suggests improvements to subject lines, length, and personalization. Integrates directly with Gmail and Outlook.
Copy.ai
Includes cold email templates and a workflow for generating variations. Useful if you are sending outreach at volume and need multiple versions quickly.
Browse AI sales and email tools in our directory
Frequently Asked Questions
How long should a cold email be?
Under 120 words is the general benchmark. Shorter is usually better. If you cannot explain the value and make an ask in 100 to 150 words, the email needs more clarity, not more words.
Should I use AI to personalize at scale?
AI can help with personalization at scale by generating customized snippets based on data you provide. However, truly personalized emails written individually still outperform bulk-personalized ones. Use AI to reduce the time cost of individual personalization, not to replace it.
What subject lines work best for cold emails?
Short subject lines of three to five words tend to outperform long ones. Questions, specific references to the recipient, and plain-language subjects outperform clever or salesy ones. Ask the AI to generate five subject line options and test the one that feels most natural.
Is it ethical to use AI to write cold emails?
Yes. You are still crafting the strategy, choosing the target, verifying the facts, and making the ask. AI helps with the writing. The ethical considerations around cold email are the same whether you write it manually or with assistance.
Summary
Cold email works when it is short, specific, and focused on the recipient. AI tools make it faster to draft, easier to iterate, and simpler to fix common problems like generic openers and weak calls to action.
The process is: define your target, write a specific prompt, evaluate the draft, refine with follow-up prompts, and add one personal detail before sending. Follow that process and your reply rates will improve.