Most advice on cold email length is a single number: "keep it under 100 words." It is a decent starting point and a poor rule. We have seen 40-word emails fall flat and 140-word emails book meetings, and the difference was never the word count itself. It was whether each sentence did a job the prospect cared about.
If you run outbound at volume, length matters for three reasons: it changes how your message reads on a phone, it affects how mailbox providers score your content, and it shapes how much work the reader must do to answer. This guide gives you a practical framework for choosing length by touch, persona, and deal size, and a clean way to test it without wrecking a working campaign.
Short emails tend to outperform long ones because they usually contain fewer irrelevant claims. Cut an email in half by deleting your company history and three feature bullets, and replies often improve. Cut it in half by deleting the one sentence that explained why you are writing to this person, and replies drop.
So think of length as the output of a decision process, not an input:
1. One reader, one problem. Name the specific situation that triggered the email. 2. One proof point. A single result or comparison that makes the claim believable. 3. One ask. Covered in detail in our post on interest asks versus calendar links.
Run those three components and you land between 50 and 120 words almost every time. When an email exceeds that range, it is usually carrying a second problem, a second proof point, or a second ask. Remove the extra and the length fixes itself.
Takeaway: Write the three components first, then count. If you are over 120 words, find the duplicate job.
Different touches in a sequence serve different purposes, so they should not share a length target. Here is the range we use as a starting hypothesis, not a law:
| Touch | Purpose | Starting range | |---|---|---| | 1 | Establish relevance, ask a low-friction question | 50-100 words | | 2 | Add one new piece of evidence | 40-80 words | | 3 | Offer something concrete (resource, comparison, short idea) | 70-130 words | | 4 | Clean break-up | 20-45 words |
Two patterns are worth noting. First, the middle touch can run longer than people expect, because by then the prospect has seen your name and a longer, more useful note feels like value rather than an imposition. Second, the break-up email should be the shortest in the sequence. A two-line close is hard to misread and easy to answer.
The mistake we see most often is copy-pasting the same 90-word structure into every step. Prospects notice. Vary the length and the angle together, and follow the structure in our cold email sequence framework so each touch earns its place.
Takeaway: Set a length range per touch, and make touch four the shortest.
A fixed range ignores who is reading. Three variables should move your target up or down.
Seniority. Executives read on phones between meetings and reward brevity. For C-level and VP recipients, aim at the low end of each range. Managers and practitioners who will personally use the product often tolerate, and sometimes prefer, more specifics, so the upper end works.
Deal size and complexity. A $3,000 annual contract can be pitched in a few lines. A six-figure deal sold to a buying committee usually needs a bit more context to show you understand the environment. That does not mean more pitch. It means more precision about the prospect's situation. For multi-stakeholder deals, see our guide to multi-threaded outreach across a buying committee, where each contact gets a message tailored to their own concern rather than one long generic note.
Familiarity with the category. If prospects already buy tools like yours, you can skip the explanation and go straight to the differentiator. If you are introducing a new category, you need a sentence or two of framing, and the email will be longer.
A worked example. Suppose you sell compliance tooling to mid-market fintech companies:
- VP of Engineering, touch one (about 55 words): A single observation about their recent audit-related job postings, one sentence on how a similar team cut evidence-collection time, and a yes/no question. - Head of Compliance, touch one (about 95 words): The same trigger, plus a specific line about the regulatory change affecting their product, and a question about who owns evidence gathering today.
Same company, same offer, two lengths, both justified by what each reader needs to decide.
Takeaway: Shorten for senior readers, lengthen only for precision, never for pitch.
Length is not a major deliverability factor on its own, but it influences the things that are.
- Very short emails with a link can look like phishing-style patterns to filters. A 25-word message whose only content is a URL is higher risk than a 70-word message with no link. - Very long emails tend to accumulate links, formatting, images, and promotional phrasing, all of which raise the risk of landing in spam or Promotions. We cover that in detail in our guide to plain text, links, and tracking. - Repetitive templates matter more than length. If thousands of messages share the same body down to the character, pattern detection gets easier. Real variation in the opening and proof lines, not just a swapped first name, protects you.
If you are scaling volume, keep the first touch plain text, link-free, and moderate in length, then monitor placement as you go. The broader foundation is in our cold email deliverability checklist.
Takeaway: Short is fine, but a short email that is mostly a link is not.
Length tests are easy to run badly. Replies are noisy, and small samples mislead. Use this process:
1. Change only length. Keep the same trigger, proof point, ask, and subject line. Create a trimmed version by removing the least essential sentence, not by rewriting. 2. Split within one segment. Send both versions to the same ICP segment on the same days, so timing and list quality do not skew results. 3. Set a sample floor. As a rough guide, aim for at least 300 to 500 contacts per variant before drawing conclusions, and more if your reply rate is low. Fewer than that, and a handful of replies will swing the result. 4. Judge by meetings held. A shorter email can lift raw replies while attracting more "not interested" responses. Compare positive replies and booked meetings, not just total responses. Our outreach metrics guide explains which numbers to track. 5. Retest by segment. A winner for founders may lose for IT directors. Treat results as segment-specific until proven otherwise.
For a broader testing structure, our A/B testing framework for cold email covers sample sizes and sequencing in more depth.
Takeaway: Test length as an isolated variable, and judge it by meetings, not replies.
Writing for the sender's ego. Company history, award mentions, and a list of integrations add length and no relevance. Cut them.
Padding the opener. "I hope this email finds you well" and "I wanted to reach out" spend your first line, which is the one the inbox preview shows, on nothing.
Cutting the trigger. The sentence that explains why you are writing now is the last thing to remove. Without it, the email reads as a mass blast.
Ignoring mobile. Preview your email on a narrow screen. Paragraphs of four or more lines become walls of text. Two to three short lines per paragraph read cleanly.
Letting automation dictate length. If a template generates variable-length output, set minimum and maximum bounds and review a sample before launch. Unbounded generation produces 200-word outliers that quietly drag down results.
Treat 50 to 120 words as the working zone for a first touch, then adjust for seniority, deal complexity, and category awareness. Vary length across the sequence, keep the final touch brutally short, and test one change at a time against meetings held.
The harder part is doing this consistently across hundreds of prospects without the quality slipping. That is where OnyxSend helps: our platform researches each prospect, drafts sequences within the length ranges you set, varies the structure per touch, and tracks which variants book meetings by segment, so you are optimizing on outcomes rather than guesses. Teams weighing this against hiring can compare the economics in our look at SDR replacement costs. When you are ready to try it on a small segment, see the OnyxSend pricing page.