Cold Email Testing Email Length: What Actually Works (With Real Data)

You're probably sending emails that are too long. Not wildly too long - just long enough that busy people skim them, miss the point, and delete them without responding.

The problem is that most advice on email length is either "keep it under 50 words" (which is impractical) or "as long as it needs to be" (which tells you nothing). Neither approach works when you're trying to actually move metrics on cold email campaigns.

Here's what we've learned from running hundreds of campaigns: email length matters less than structure, but structure changes how long your email needs to be. And there's a specific range where most B2B cold emails perform best.

The Benchmark: What Length Actually Works

Across B2B cold email campaigns we've tracked, the sweet spot sits between 75 and 150 words. Not 50 words. Not 250 words. This range.

Here's why: emails under 75 words often feel hollow. You don't have enough space to reference something specific about the prospect without sounding generic. Emails over 150 words lose people who are reading on mobile or scanning quickly - which is most of your inbox.

But here's the nuance: the word count matters less than what those words do. A 120-word email that's poorly structured performs worse than a 140-word email with clear line breaks and one real reason to respond.

We tested this directly on a SaaS sales consulting client. We took two emails:

Email A: 68 words, single paragraph, no personalization hook

Email B: 118 words, three short paragraphs, one specific company reference

Email B outperformed Email A by 34% in reply rate across 4,000+ sends. The extra 50 words weren't filler - they did actual work.

The Structure That Lets You Use More Words Without Losing People

The length that works depends entirely on how you structure it. Three short paragraphs at 120 words reads faster than one dense paragraph at 80 words.

Here's the framework that works:

Paragraph 1 (1-2 sentences): One observation about them or their situation - something that proves you didn't mail-merge this
Paragraph 2 (2-3 sentences): One specific thing you do and one reason it matters to their situation
Paragraph 3 (1 sentence): A low-friction ask

This structure lets you hit 110-130 words without it feeling long because the white space makes it scannable. Most people don't count words - they feel length. Structure controls how long it feels.

Here's a real example from an agency we worked with:

Hi [First Name], Saw you brought on [New Team Member] in a dev lead role - good timing since most dev teams are drowning in technical debt before Q4 sprints. We work with agencies like [Competitor] to untangle legacy code and get teams back to shipping features instead of firefighting. Usually takes 3-4 weeks to show where the real bottlenecks are. Worth a quick call? [Name]

That's 74 words. Feels short because of the structure, not long because of the personalization doing work. The point isn't the word count - it's that every sentence earns its space.

When Length Actually Matters (And When It Doesn't)

Email length only impacts performance when your message is unclear. If someone knows what you want in the first three sentences, they'll respond or they won't - extra words don't change that.

Where length kills you:

When you explain your solution before explaining why it matters to them
When you use 40 words to say something that takes 10
When your ask is buried in paragraph 3 and not obvious until people re-read

Where length doesn't hurt you:

When every sentence references something specific about their business
When you're showing competence through a relevant example, not bragging
When the structure makes it easy to scan (short paragraphs, one idea per sentence)

We tested this on a web design agency campaign. We had two versions of their email, both 95 words:

Version 1: Heavy on their process and credentials. Vague ask at the end.

Version 2: Opened with a specific observation about the prospect's site. One sentence about their approach. Clear ask.

Version 2 hit 28% reply rate. Version 1 hit 12%. Same length, different structure, different outcome.

The Length Test You Should Actually Run

Don't test "short vs. long." That's useless because you don't know which element is causing the change. Instead, test length in fixed 25-word bands while keeping structure constant.

Split your audience into four groups:

Group A: 75-100 words
Group B: 100-125 words
Group C: 125-150 words
Group D: 150-175 words

Run 500-1000 sends to each group. Track reply rate, not open rate (length doesn't change opens much, but it changes reply behavior).

When you do this across different industries, Group B and C usually win. But "usually" isn't the same as "always." Your specific audience might prefer something different - that's why you test, not guess.

For a managed IT services company we worked with, Group D (150-175 words) actually outperformed. Why? Their decision-makers were CIOs who expect depth. They weren't skimming - they were evaluating. Length worked in that context.

For a fractional CFO service, Group A (75-100 words) won by 18%. CFOs are time-constrained and want the ask immediately. Same industry vertical, opposite result.

The Real Constraint: Are You Writing Long or Are You Writing Badly?

Here's the conversation we have with clients: "Is your email long, or is it just inefficient?"

There's a difference. A 140-word email that uses every word strategically converts better than a 90-word email padded with fluff. But most "long" emails are just poorly written.

Before you test length, audit your current email. Go through sentence by sentence and ask: "Does this sentence make someone more likely to respond, or am I just filling space?" If it's filler, cut it. If it's substance, keep it.

Here's an example of what to cut:

Hi [First Name], I hope this email finds you well. I've been in the industry for 15 years and have seen a lot of changes. We work with companies like yours to help them with their challenges. I think we could be a good fit. Would you be open to a conversation? [Name]

That's 66 words of nothing. Every sentence is generic. The length isn't the problem - the content is.

Now the same email rewritten to actually work:

Hi [First Name], Most compliance teams we work with are running audits 3-4 weeks after they should because tools aren't talking to each other. We built a connector that syncs your audit trail to your compliance dashboard in real time. Worth 20 minutes to see how other teams are handling this? [Name]

That's 56 words and it works because every word does something. It's not about hitting a word count - it's about hitting a conversion rate.

Once you understand what you're actually trying to communicate, the length usually takes care of itself. You'll write what you need to write. Then test variations around that baseline.

A Practical Next Step

If you want to test length on your own campaigns, start here: write three versions of your email at 80 words, 120 words, and 150 words. Keep the core message identical - just add specificity and examples in the longer versions, not fluff.

Split your list evenly and send. Track replies for two weeks. That single test will tell you more about your specific audience than any benchmark.

Then use that result to inform your ongoing cold email split testing. Length is one variable - but it's one that most people guess on instead of measure.