Write emails with AI that don’t sound automated (even if they are)

Course content

Resources

Most so-called “personalized” emails still sound generic, and prospects can sense it immediately. That’s why so many end up ignored or deleted.

But with the right approach, you can use AI to craft emails that feel genuinely tailored, spark authentic interest, and actually get replies. In this guide, we’ll cover four proven methods to write highly personalized AI emails that open doors (not just inboxes).

Let’s dive in.

How to write personalized sales emails that convert

Let’s start with the fundamentals: email is far from dead. It’s not even declining. In fact, email remains one of the most reliable, scalable, and measurable channels in B2B sales.

Why? Because it meets your prospects where they already work. Unlike social DMs or cold calls that interrupt, emails can be opened, forwarded, saved, or replied to on their schedule.

And yet, most sales emails still look like they were written by someone who didn’t bother to learn a single thing about the person they’re contacting. That’s why response rates hover in the low single digits for many teams.

But here’s the encouraging news: by applying a few essential principles, you can increase your chances of hearing back. Even with cold emails. Even at scale.

Oscar-Wyett Moore puts it this way:

It is possible to achieve response rates of up to 80% even with AI-written emails.

Oscar-Wyett Moore – Director of User & Customer Experience @ Scratchpad

That’s not a typo, but it only happens when you get the fundamentals right.

The non-negotiables of personalized emails that actually convert

1. Address a genuine pain point (or lose your prospects instantly)

You can have the most compelling subject line in the world, but if your email doesn’t speak to an authentic pain or priority your prospect is facing, it’s destined to be ignored.

That means your personalization has to go beyond “Hey [first name], saw you went to [first name]” The only personalization that matters is around context and problems*.*

So how do you find those?

Use Intent Data:

Tools that track content consumption, tech stack changes, hiring trends, or even funding rounds can give you signals about what might be top of mind for your target.

Look at role-specific challenges:

A VP of Sales and a Head of Customer Success care about different metrics. Your message should reflect that.

Mirror the language they use:

Scan their LinkedIn posts, company blogs, or press releases. Borrow their wording to prove you understand their world.

When your email pinpoints a problem they actually lose sleep over, you’ve already won half the battle.

2. Deliver value immediately

No one wakes up hoping to read more sales emails. If you’re going to interrupt their day, make it worthwhile.

That means leading with value, not your product.

Ask yourself: Why should they keep reading after the first two lines?

Ways to show immediate value:

Share a brief, relevant insight (like a metric or benchmark from a similar company).
Offer a hypothesis of what they might be facing: “Noticed your team doubled headcount last quarter, often that’s when onboarding consistency starts to slip. Seeing that?”
Pose a thoughtful question that frames the problem in a new light.

The biggest mistake? Saving the hook for the end. Don’t bury your value under paragraphs of who you are. They don’t care (yet).

3. Make the next step clear and low friction

When someone gets to the end of your email and thinks “So… what do you want from me?”, you’ve lost.

Every effective sales email does three things by the close:

It offers a clear next step.
Usually a brief call, but could also be a piece of content or even just a yes/no question.
It makes saying yes easy.
Suggest times, offer to work around their calendar, or use a simple “worth exploring?”.
It ties back to the problem.
Reinforce why taking that step matters.

This is how you help the prospect visualize the logical progression: “This is your challenge, here’s why it matters, and here’s how we can start solving it together.”

Can you still personalize at scale? (Yes, with a few strategic plays)

The reality is you probably have hundreds or thousands of leads to reach out to. That’s where tools and AI come in. The danger is sliding back into generic mass outreach. But with the right approach, you can keep the personal touch intact.

Here’s how:

Segment strategically.
Break down your leads not just by company size or industry, but by common challenges or triggers. Each segment gets messaging tailored to that theme.
Use dynamic fields beyond name and company.
If your enrichment data includes tech stack, recent funding, or hiring patterns, you can weave that in to make emails feel handcrafted.
Layer human review.
AI can draft hundreds of personalized lines, but have your reps sanity-check and refine them. A quick 10-second human pass often turns a decent line into an excellent one.

The result? You get the efficiency of automation without losing the credibility of genuine insight.

3 methods to write highly personalized AI emails that get replies

Method #1: AI as a writing assistant

This might surprise you: the best AI-powered emails usually don’t come from letting ChatGPT or another model write everything from scratch. According to Matthew Ray Scott (Chief Evangelist at Virtual Sales RX) they come from treating AI like a sparring partner, a smart assistant who helps you sharpen your messaging, stress-test it, and push it until it’s truly compelling.

How it works

Write a draft by hand.
Start by putting down your rough idea. The point is to get your thoughts out, your positioning, the pain point you want to address, maybe a hook you like.
Feed AI rich context.
Give it more than just “write me a sales email.” Upload successful past emails, examples that worked, info about your company, the prospect’s business, your objectives, and specific challenges.
Have AI clean up your V1.
Ask it to rewrite your draft into a clear, concise first version. This gets rid of fluff and polishes your language.
Switch roles: let AI act as your prospect.
Prompt it: “You’re Emily, an HR manager at a tech company moving to hybrid. How would you respond to this? Be brutally honest.”
This reveals what might land awkward, unclear, or too self-serving.
Iterate, refine, repeat.
Use the feedback to tighten your message. Then ask AI to analyze again. Try multiple angles: shorter vs. longer, more casual vs. more direct.
Test small variants.
Adjust CTA friction, clarify benefits, swap out generic words for sharper ones. Each pass should get closer to a message that feels truly personal.
Pick your winner.
Finally, select the version that’s most likely to earn a quick “Sure, let’s chat.“

Example prompt to feed your LLM

Context:

“You’re a business development lead at a SaaS startup with a project management platform tailored to hybrid teams. You’ve spotted a LinkedIn post from Emily, an HR manager, about her company’s new hybrid policies.”

Goal:

“Start an informal conversation, learn about their challenges, eventually offer a free trial if it’s relevant.”

Main challenge:

“Don’t come off like a typical pushy salesperson.”

Version 1 (before iteration):

Subject: Exploring hybrid work tools for [Company Name]

Hi Emily,

I recently saw your LinkedIn post about your new hybrid work policies. At [MyStartup], we help teams like yours manage projects across remote and onsite environments.

Would you be open to a quick call next week to see if our platform might fit?

Best,

Version 2 (after using AI as recipient & refining):

Subject: Quick exchange? Insights on hybrid team management

Hi Emily,

Really appreciated your LinkedIn post on how [Company Name] is evolving hybrid work, inspiring to see how you’re prioritizing both flexibility and employee well-being.

I’m reaching out not to pitch, but because we’ve been working with several tech teams tackling similar hybrid challenges, especially keeping engagement and visibility high for all team members.

Would you be open to a brief, informal call (or even just an email exchange)? I’d love to hear how you’re approaching onboarding and team cohesion, and happy to share a couple of practical tips we’ve seen work elsewhere, no strings attached.

If now’s not ideal, totally fine. Just let me know what suits you best.

Thanks so much,

What changed?

The tone shifted from “asking a favor” to a genuine peer conversation.
Generic “chat” replaced by more precise language that feels natural for HR.
CTA softened, giving Emily control over next steps.
Clear Intention to learn and add value, not just sell.

Method #2: Build custom prompts to triple your response rates

If you want AI to deliver more than generic, you need to teach it exactly what excellence looks like. The best teams do this by developing hyper-targeted prompts, backed by proven data.

How it works

Feed your AI a knowledge base.
For example, copy-paste 100+ successful emails or articles from experts like WillGPT into your assistant. Let the model “learn” from proven high-performing copy.
Use multiple LLMs.
Don’t limit yourself to one. Run your inputs through ChatGPT, Claude, and Jasper, then compare how each structures the outreach. Ask each for pros & cons.
Run long-term A/B tests.
Send these variants over months, track open, reply, and meeting rates. Build a library of top performers.
Craft your master prompt.
Once you know what resonates, write an ultra-specific master prompt that consistently delivers concise, tailored, high-response emails.

Example prompt

“Write a concise, personalized cold email to [audience: e.g., medical tech sales managers].

Start with an intriguing subject line (under 50 characters).
Open with a clear observation showing research on their challenges.
Present a direct benefit tied to their pain.
Add a mini case study.
Keep it under 100 words.
Simple sentences, friendly, no pressure close (like “Worth discussing?”).

Aim for readability score below 5.”

Resulting email:

Subject: Triple surgeon consults

Hi [First Name],

What if your team could triple consults with surgeons using a strategy that grabs attention 76% of the time?

Over 20,000 medtech reps, from J&J to Medtronic, now rely on one-minute personalized videos to cut through. The result: an average 76% engagement, even with the busiest specialists.

Happy to share how this might fit your team. Worth a quick look?

Why it works:

Value shows up early, credibility is built quickly, and the CTA is light, easy to say yes.

Method #3: Scrape real data to avoid generic "mid emails"

Here’s the harsh truth: most emails fail because they’re lazy. They talk vaguely about “growth,” “scaling,” or “alignment,” with zero substance tied to the prospect’s reality. That’s why Alex Berman (Founder and CEO of Galadon) builds mini-agentic workflows that scrape hyper-relevant data to personalize at scale.

How it works

Start by scraping the prospect’s site.
Use tools like n8n + Serper to pull in company descriptions, product pages, press releases.
Extract key details.
Have your agent auto-identify their offer, ICP, recent initiatives.
Map to your template library.
AI then picks the best-fit template from your proven set.
Layer in target audience + pain points.
AI checks for mentions of who they sell to or their main goals.
Add a micro case study.
Agent matches a customer story similar to the prospect.
Output a hyper-tailored email.

Example in practice

Hey [Name], huge fan of what you’re doing in marketing & sales. I help teams like yours level up LinkedIn with live workshops and private communities for continuous collaboration.

For instance, we worked with Kirsha Campbell of The Cash Lab to overhaul her LinkedIn strategy, drove serious business growth.

Want me to send a few ideas?

Why it’s better:

It feels like you spent hours studying them, even though it was largely automated. As Alex Berman puts it:

It can look very complicated, but it’s simple. We’ve all seen lazy, spray-and-pray fail. Build your agentic workflow, and your emails will finally stand out.

Alex Berman – Founder and CEO @ Galadon

Masterclass

I built an AI that writes perfect cold emails – Video originally published by Alex Berman

La Growth Machine: equipping yourself with a solution with integrated AI

You’ve established the right process, now it’s time to scale it efficiently. Modern AI capabilities can transform how you approach personalization, turning what used to be a time-intensive manual process into a streamlined workflow. With features like Magic Message, La Growth Machine demonstrates how AI can help you create highly personalized outreach in under 2 minutes while maintaining message quality.

Why AI-powered personalization delivers results

Accelerates message creation without sacrificing quality.
Generate new messages that maintain your strategic approach while saving time.
Enhances existing content iteratively.
Refine and improve drafts based on your specific messaging goals.
Scales personalization systematically.
Apply variables and customization across your entire outreach sequence.

Strategic implementation approach

Describe your product/ company name
Describe your product
Describe your key benefits
Describe your buyer persona
Let LGM’s AI build a message that tightly aligns your solution to their reality.

This approach ensures your personalization efforts remain both efficient and strategically sound, helping you achieve higher response rates through better message-to-market fit.

Perfect for boosting reply rates in multichannel sequences. And because it’s trained specifically for this sales workflow, it knows how to merge concise, punchy language with authentic personalization.

Takeaways

So next time you think “AI emails,” don’t think “generic.” Think smarter, sharper, more human, just massively faster.

Because when you combine AI’s horsepower with your expertise, grounding each message in real pain points, adding value upfront, and making the next step ridiculously simple, you’re no longer blasting cold emails. You’re opening warm doors.

And that’s where the real growth happens.

Write emails with AI that don’t sound automated (even if they are)

How to write personalized sales emails that convert

The non-negotiables of personalized emails that actually convert

1. Address a genuine pain point (or lose your prospects instantly)

2. Deliver value immediately

3. Make the next step clear and low friction

Can you still personalize at scale? (Yes, with a few strategic plays)

3 methods to write highly personalized AI emails that get replies

Method #1: AI as a writing assistant

How it works

Example prompt to feed your LLM

Method #2: Build custom prompts to triple your response rates

How it works

Example prompt

Method #3: Scrape real data to avoid generic "mid emails"

How it works

Example in practice

La Growth Machine: equipping yourself with a solution with integrated AI

Takeaways

Understanding AI for sales

Prospect Smarter: AI-Powered segmentation, enrichment & activation

Message better: personalization, sequencing and conversion

Qualify and close: smarter calls, faster conversions