Email Marketing Copy Guide: Subjects, Preview, Body & CTAs

Email is still the highest-ROI channel in marketing — but only when the copy works. Subject lines decide opens, preview text decides hesitation, body decides intent, and CTAs decide revenue. Each piece is a separate craft, and ignoring any one of them undercuts the rest.

Email Copy Component Lengths

ComponentIdeal LengthPurpose
Subject line30-50 charsEarn the open
Preview text50-90 charsReinforce, intrigue
Opening sentenceUnder 12 wordsKeep them reading
Body (promo)50-125 wordsOne idea, one ask
Primary CTA2-5 wordsGet the click

Subject & Preview

Treat them as a pair. The subject is a question or hook; the preview is the down payment on the answer. Avoid restating the subject — that wastes the second hook. Strong combinations: "Your June numbers" / "Three insights inside (avg. read: 2 min)."

Body Structure

  1. Opening — context in one sentence. No "Hope you're doing well."
  2. Value — the offer, story, or insight. Lead with the benefit.
  3. Proof — a number, name, or quote when relevant.
  4. CTA — one clear action, button-styled if possible.
  5. PS line — second-most-read line; reinforce urgency or add a bonus.

CTAs That Convert

Use specific verbs and first-person framing where appropriate. "Start my free trial" outperforms "Start free trial" because it primes ownership. Keep CTAs benefit-led: "Get my discount" beats "Submit." Place the primary CTA above the fold and repeat once more before the signature.

Test Your Subject Lines

Score subject lines for length, sentiment, spam triggers, and emoji impact.

Email Subject Tester →

Subject line: 41 characters or fewer

Mobile clients truncate subject lines around 41–50 characters depending on inbox app and orientation. Litmus' 2024 inbox report puts the median open-rate-maximising length between 28 and 50 characters. Front-load the value: "Your invoice is ready" beats "A quick update from the team about your account". Personalisation tokens still help when used sparingly — first-name in the subject lifts opens 6–10% on average but loses effect if every brand does it.

The preheader is half your real estate

Most teams paste the first sentence of the body into the preheader by accident. That sentence is almost always a greeting ("Hi Sarah,") and wastes the strongest piece of post-subject real estate. Write the preheader as a deliberate continuation of the subject: subject "Your invoice is ready", preheader "Pay by Friday to keep this month's discount". Together they should answer "why open this?" in a glance.

Body structure that actually gets read

  1. One thing per email. If you have three asks, send three emails or a single nurture sequence.
  2. Front-load the ask. The button should be visible in the first 100 pixels. People skim — they do not scroll.
  3. Short paragraphs. One to three sentences. Long blocks read as marketing noise even when they are not.
  4. One primary CTA, max one secondary. Multiple equal-weight buttons reduce click-through, they do not increase it.
  5. Plain-text fallback. A meaningful version that is not "View this email in your browser" — Gmail Promotions weights heavily-imaged emails down.

Worked rewrite

Before: "Hello! We hope this email finds you well. We wanted to let you know that we have just launched our new product and we would love it if you could take a look and let us know what you think. There are lots of new features and we think you will love them."

After: "We shipped three things you asked for: dark mode, CSV export, and Slack integration. Open the changelog — five-minute read."

Send-time matters less than you think. A/B test inside your audience. The "best send time on Tuesday at 10 am" rule is a 2014 artefact; modern inbox algorithms have largely flattened the curve.

Frequently Asked Questions

30-50 characters with the keyword front-loaded for mobile truncation.
Yes — it's the second headline. Write it manually, don't let it default.
Promo: 50-125 words. Newsletter: 200-300. Always one primary ask.
One primary CTA, repeated 2-3 times pointing to the same destination.
Sparingly. Test per audience — lifts B2C, can hurt B2B.