Skip to main content
Academy · Module 4 of 9

Writing emails people answer

What the published studies say about length, subject lines and calls to action, and what they don't. Relevance beats decoration, and one low-commitment ask beats a calendar link.

14 min read · Course overview

Objectives

In this module

  • Keep first emails short, and know what the length studies do and don't show.
  • Write plain, honest subject lines.
  • Tell relevance apart from decorative personalization.
  • Match the call to action to how warm the relationship is.
  • Use writing frameworks as scaffolds, not formulas.
4.1

Short wins, modestly

Several large datasets point the same direction. Hunter's analysis of about 34 million emails found 20–39 words had the highest average reply rate, at 4.5% [1]. Smartlead's 2026 data shows first emails under 50 words needing fewer contacts per reply than those of 100–149 words [2]. Gong found 30–150 words worked best for booking meetings [3]. Instantly observes that its top performers keep first emails under 80 words [4].

Read these carefully. The differences between length bands are often small, Hunter's own author says word count alone is unlikely to move replies much [1], and all of it is platform data rather than experiment. The useful lesson is that short emails force you to say one thing clearly. Often-repeated claims about exact word counts or reading grade levels come from material without a published method; don't build on them.

4.2

Subject lines

Woodpecker found subject lines of 20–29 characters had the best reply rates in its Q4 2025 data [5]. Short and plain, the way a colleague writes: “Question about Q1 hiring”, not “🚀 Unlock 10x growth”.

Two rules aren't optional. CAN-SPAM forbids deceptive subjects [6], so a fake “Re:” or “Following up on our call” on a first email is out. It's also exactly the kind of fake one-to-one signal M3AAWG criticizes [7]. And because Apple preloads images, you can't judge subject lines by open rate; judge them by replies (Module 8).

4.3

Relevance, not decoration

No benchmark cleanly isolates the effect of personalization. lemlist's 2026 review notes that first-name-and-company merge fields read as automated, and that top performers rely on tight targeting and timing more than personalization alone [8].

Distinguish relevance (why this person, why now, what problem) from decoration (a compliment about their latest post). A first line that names a real trigger does both. A flattering line that could be sent to anyone does neither, and recipients can tell.

Decoration vs. relevance
Decoration:
  “Loved your recent post on leadership! Hope you're crushing Q4.”

Relevance:
  “Saw you're hiring two SDRs in Austin. Teams adding reps usually
   hit the same snag: nobody owns which prospects have already
   been emailed, by whom.”
4.4

One ask, sized to the relationship

Gong Labs analyzed 304,174 emails and found that on cold first emails, an interest-based ask (“Is this something you're open to exploring?”) outperformed asking for a specific meeting time [9]. Once a deal was live, the specific-time ask won: 37% against 25% for interest-based asks [9]. The data is from 2020, so treat it as direction rather than exact.

The principle: match the size of the ask to how warm the relationship is. A stranger can easily answer “worth a look?” but has to work to answer “Tuesday at 2pm?”. Use one call to action per email [4]. Two questions halve the chance of an answer to either.

A short first email
Subject: SDR hiring in Austin

Hi Dana,

Saw you're hiring two SDRs in Austin. Teams adding reps usually hit
the same snag: nobody owns which prospects have already been emailed,
by whom, and what they said.

We built a shared inbox and do-not-contact list for exactly that.
Worth a look, or not a priority right now?

Will

—
AngelSend · 123 Example St, Austin TX · Not relevant? Unsubscribe
4.5

Format and frameworks

No published study compares plain text with HTML for cold B2B email. The common practice is an email that looks one-to-one: minimal formatting, few or no links, no images. Whatever the format, a commercial email still needs its postal address and opt-out [6].

Frameworks help you structure a draft. AIDA (attention, interest, desire, action) suits longer nurture emails. PAS (problem, agitate, solution) suits pain-led emails, though “agitate” can easily read as manipulative. BAB (before, after, bridge) suits outcome-led emails. For short first emails, question–value–ask works well. None has been shown to outperform the others, so use them to get to clarity, then cut.

Key takeaways

Remember this

  • Short first emails are a good default; exact word counts are not law.
  • Subject lines: short, plain, honest. Never fake a reply thread.
  • Relevance (why you, why now) beats flattering personalization.
  • Cold: ask about interest. Warm: ask for a specific time.
  • One call to action per email, plus the legally required footer.
Exercise

Rewrite one email

Take the last cold email you sent, or the template you use most.

  1. Step 1

    Cut it to under 100 words without losing the reason for writing.

  2. Step 2

    Replace any decorative opener with one sentence of real relevance, or delete it.

  3. Step 3

    Reduce it to one call to action, and make it interest-based if the recipient doesn't know you.

  4. Step 4

    Run it through a spam checker and check the footer has your postal address and an opt-out.

In AngelSend

Where this lives in the product

Spam check

Campaign copy gets a spam audit before it sends, and the free spam checker at /tools/spam-checker runs the same kind of check on any draft.

Subject line drafts

The assistant can suggest subject line options and email variants from your campaign brief for you to edit.

Merge fields and conditional blocks

Merge variables and {{#if}} blocks let one template say something different, and true, for contacts with and without a given field.

Check yourself

Quiz

Pick an answer to see the explanation.

  1. Q1On a cold first email, which call to action did Gong's data favor?

  2. Q2Which subject line is acceptable on a first cold email?

  3. Q3What's the right way to read “20–39 words had the highest reply rate”?

Sources

  1. [1]What's the Best Word Count & Length for Cold Emails — Hunter
  2. [2]The State of Cold Email 2026 — Smartlead
  3. [3]Cold email stats — Gong
  4. [4]Cold Email Benchmark Report 2026 — Instantly
  5. [5]Cold Email Benchmarks — Woodpecker
  6. [6]CAN-SPAM Act: A Compliance Guide for Business — U.S. FTC
  7. [7]M3AAWG Position on Cold Email — M3AAWG
  8. [8]Cold Email Benchmarks 2026: What Good Actually Looks Like — lemlist
  9. [9]This surprising cold email CTA will help you book a lot more meetings — Gong Labs

This course summarizes published rules, regulator guidance and third-party research as of October 2026. It is not legal advice; check with counsel before emailing into a jurisdiction. Vendor benchmark figures are each vendor's own data and definitions.

3. Who to email5. Follow-ups and sequences

Practice on
your own outreach.

Start with a $3.99 Starter Pack — no subscription required.

Get started freeBook a demo

Starter pack from $3.99 · CAN-SPAM compliant · Reputation protected