How to Write a Cold Email That Gets Replies (Data-Backed)

Templates don’t work. Patterns do.

Every cold email guide gives you templates. “Just swap in your company name!” Then 500 other people send the exact same email, and reply rates crater.

Templates expire. Patterns don’t. This guide focuses on the structural decisions backed by data — the length, format, and psychological triggers that consistently drive replies, regardless of what you’re selling.

The numbers here come from analysis of cold email campaigns across multiple platforms and published research from Lavender, Gong, Mailshake, and our own data at SendEmAll.

The numbers that matter

Email length: 50-125 words

This is the single highest-impact variable you control. Cold emails between 50-125 words consistently outperform longer messages.

Word Count	Average Reply Rate
Under 50 words	5-8% (too vague, seems automated)
50-75 words	10-14%
75-125 words	11-15% (sweet spot)
125-150 words	8-11%
150-200 words	6-9%
200+ words	3-6%

Reply rates drop roughly 15% after 150 words. Every word past 125 needs to earn its place.

Why? Your prospect didn’t ask for this email. They’re scanning it on a phone between meetings. You have 8-11 seconds of attention. That’s about 75 words read at average speed.

Subject lines: 3-5 words beat 8+

Short subject lines win in cold email. They look like internal emails, not marketing blasts.

Subject Line Length	Open Rate Impact
1-2 words	High opens, but can feel spammy
3-5 words	Highest open rates (+18% vs average)
6-8 words	Average performance
9+ words	Looks like marketing, -12% opens

Personalized subjects (containing the prospect’s company name or a specific pain point) boost open rates by 22-26% compared to generic ones.

Lowercase outperforms Title Case by 5-8% in cold email. Lowercase reads as casual and personal. Title Case reads as promotional.

Good: quick question about [[company]] Bad: Quick Question About Your Marketing Strategy for Q2

“Re:” and “Fwd:” tricks — these worked in 2022. In 2026, Gmail and Outlook flag them as deceptive if there’s no actual thread. Don’t risk it.

Emojis — they hurt B2B cold email open rates by 3-5%. Save them for consumer marketing.

The first line: make it about them

The first sentence determines whether someone reads the rest. Here’s what works and what doesn’t.

What kills reply rates:

• “I hope this email finds you well” (adds nothing, signals template)
• “My name is [Name] and I work at [Company]” (they don’t care yet)
• “I noticed you’re the [Title] at [Company]” (so did the other 20 people who emailed today)

What drives replies:

• Reference something specific: a recent company announcement, a job posting, a tech stack choice
• Name a problem they’re likely experiencing based on their role or company stage
• Connect a signal to your outreach reason

Examples of strong first lines:

• “Saw you’re hiring 3 SDRs — scaling outbound while keeping deliverability high is a real challenge at that stage.”
• “Your team just switched to HubSpot from Salesforce — curious how the data migration went on the outbound side.”
• “Congrats on the Series B. Most post-funding companies triple their outbound volume and immediately hit inbox placement issues.”

Each one shows you did homework. Each one connects to a problem you can solve. That’s the difference between 3% reply rates and 15%.

The value proposition: one sentence, specific outcome

You get one sentence to explain why they should care. Not a feature list. Not your company story. One specific outcome they’d want.

Weak: “We offer an AI-powered outbound platform with email verification, enrichment, and multi-channel sequencing.”

Strong: “We help B2B sales teams reach 200+ signal-qualified potential buyers per month for $149 — including the infrastructure, data, and personalization most tools charge separately for.”

The strong version has a number (200+), a qualifier (signal-qualified), a price ($149), and a differentiator (everything included). Four concrete data points in one sentence.

The CTA: one question, low commitment

The single biggest CTA mistake: asking for too much too soon.

High friction (low reply rate): “Would you be available for a 30-minute demo next Tuesday?”

Low friction (high reply rate): “Worth a 15-minute call to see if this fits?”

Even better — ask a question that doesn’t require a time commitment:

• “Is outbound a priority for you this quarter?”
• “Would it be useful to see how this works for [their industry]?”
• “Mind if I send a 2-minute walkthrough?”

One CTA per email. Not two. Not “let me know either way.” One clear, low-commitment question.

Follow-up cadence: 3-5 touches, spaced correctly

Most replies come on follow-ups, not the initial email. But timing matters.

Touch	Timing	Reply Rate
Email 1	Day 0	4-6%
Follow-up 1	Day 3	3-5% (cumulative: 7-11%)
Follow-up 2	Day 7	2-4% (cumulative: 9-15%)
Follow-up 3	Day 12	1-3% (cumulative: 10-18%)
Follow-up 4	Day 21	1-2% (cumulative: 11-20%)

Each follow-up must add new value — a different angle, a case study, a relevant data point. Never just “bumping this up” or “circling back.” Those get deleted instantly.

For more on follow-up sequences, see our detailed guide: Cold Email Follow-Up Sequence: How Many, When, and What to Say.

Three frameworks that work

1. PAS (Problem - Agitate - Solution)

Structure:

• Problem: Name a specific pain they have
• Agitate: Show the consequence of not solving it
• Solution: Your one-sentence value prop + CTA

Example:

Subject: outbound costs

Hi [[firstName]],

Running outbound with separate tools for leads, verification, sending, and infrastructure typically costs $400-600/month before you send a single email.

That budget pressure usually means teams cut corners — skip verification, under-invest in domains, or blast from too few mailboxes. All of which tank deliverability and waste the money you did spend.

SendEmAll bundles everything into one platform starting at $149/month. Data, verification, sending, managed infrastructure.

Worth a quick look?

Word count: 82. Names a specific dollar problem. Agitates with real consequences. Offers a specific alternative.

2. AIDA (Attention - Interest - Desire - Action)

Structure:

• Attention: A hook tied to something relevant to them
• Interest: A fact or insight they didn’t know
• Desire: What they’d get
• Action: Low-friction CTA

Example:

Subject: [[company]] outbound

Hi [[firstName]],

Noticed you posted an SDR role last week — looks like you’re scaling outbound.

Most teams at your stage spend 3-4 weeks just setting up domains, mailboxes, DNS records, and warmup before sending a single campaign. That’s a month of your new hire sitting idle.

We handle all of that. Your team sends from day one with managed infrastructure, verified potential buyers, and AI-drafted personalization they approve before it goes out.

Would a quick walkthrough be useful before your new SDR starts?

Word count: 97. The hiring signal makes it timely. The “month of idle time” pain is specific. The CTA ties directly to their situation.

3. Before-After-Bridge (BAB)

Structure:

• Before: Their current painful state
• After: What life looks like with the problem solved
• Bridge: How to get there

Example:

Subject: quick question

Hi [[firstName]],

Right now, your outbound stack probably looks like this: one tool for leads ($99/mo), another for verification ($50/mo), a sending platform ($97/mo), plus domains and mailboxes ($100/mo). Total: ~$350/month, and you’re still managing integrations between all of them.

What if all of that was one platform, one bill, one dashboard — at $149/month? Including signal-qualified potential buyers, not just scraped contacts.

That’s what we built. SendEmAll replaces 4-5 tools with one, and your emails actually land because we manage the infrastructure end to end.

Want to see how it compares to your current stack?

Word count: 118. The “before” uses real dollar amounts. The “after” is tangible. The bridge is the product.

Before vs after: the same email, rewritten

Before (generic, 187 words):

Subject: Introducing Our Sales Engagement Platform

Hi there,

I hope this email finds you well. My name is Alex and I’m reaching out from SendEmAll. We’re a comprehensive outbound sales platform that helps businesses streamline their sales development process.

Our platform offers a wide range of features including AI-powered personalization, email verification, lead enrichment, multi-channel sequences, and advanced analytics. We’ve helped hundreds of companies improve their outbound results.

I’d love to schedule a 30-minute demo to show you how our platform could benefit your team. We have availability next Tuesday or Wednesday afternoon.

Looking forward to hearing from you!

What’s wrong: Generic greeting. Self-focused intro. Feature dump. No specificity. High-friction CTA. 187 words. Would get 1-2% reply rate.

After (specific, 89 words):

Subject: [[company]] outbound costs

Hi [[firstName]],

I looked at your tech stack on BuiltWith — you’re running HubSpot, Outreach, and Apollo. That combination usually runs $400-700/month for a team your size.

We consolidate all of that into one platform at $149-349/month. Same volume, better deliverability, and every potential buyer is signal-qualified before you email them.

Your Apollo contract likely renews in Q3. Worth comparing numbers before then?

— Alex

What’s right: Personalized subject. Specific tech stack reference. Real dollar comparison. Timely CTA tied to their renewal cycle. 89 words. Would get 10-15% reply rate.

The writing checklist

Before you send any cold email, check these:

Check	Target
Word count	50-125 words
Subject line	3-5 words, lowercase, personalized
First line	About them, not you
Value prop	One sentence, specific number or outcome
CTA	One question, low commitment
”I” count	Fewer than “you” count
Follow-up plan	3-5 touches with new value each time
Personalization	At least one specific reference to their company

How SendEmAll helps you write better emails

The data patterns above are what SendEmAll’s AI personalization is built on. The platform:

• Monitors buying signals (hiring, funding, tech changes) and surfaces them for your first line
• Generates first-draft emails using signal context — not generic templates
• Keeps emails within the 50-125 word sweet spot
• Writes subject lines based on the prospect’s specific situation
• You review and approve everything before it sends

You bring the strategy and voice. The AI handles the research and first drafts. The result is personalized outbound at scale that actually sounds like a human wrote it — because a human approved it.

Start writing better cold emails. 100 free credits to test the approach.