What this is for#
Subject lines have ~3 seconds to earn the open. The good news: the patterns that work are well-known + repeatable. You don't need to be clever every time. This guide is the 12 formula families that consistently outperform random subject lines, with worked examples + the A/B testing protocol to validate them on your own audience.
This is the practical companion to Email Preheader Text and A/B Testing Email Subject Lines — those cover the theory of how subjects + preheaders work together. This article gives you the patterns to start from.
The 12 formula families#
1. The curiosity gap#
Open a loop, don't close it.
- "What 3M sent that broke their email program for a week"
- "Why we deleted half our subscriber list (and got better results)"
- "The one Gmail setting that explains your open rate drop"
Why it works: the brain hates unresolved loops; opening solves the itch. Don't bait-and-switch — the email body must deliver on the loop or you teach subscribers to ignore you.
2. The specific number#
Round numbers feel like estimates; specific numbers feel like data.
- "How we increased open rate from 14% to 26% in 6 weeks" (better than "How we improved our open rate")
- "7,243 subscribers we removed last month" (better than "Subscribers we cleaned")
- "Why 47% of our new subscribers never opened a second email" (better than "Why subscribers churn")
Why it works: specificity signals real data behind the claim.
3. Personal first-name hook#
Use {FIRST_NAME} — but lightly.
- "{FIRST_NAME}, your March numbers are in"
- "Quick question, {FIRST_NAME}?"
- "{FIRST_NAME} — we noticed something about your account"
Why it works: seeing your own name triggers attention. Caveat: overuse cheapens it. Don't use first-name in every subject line — keep it for moments that genuinely warrant personalization.
4. Status / FOMO#
"Something is about to be true that you don't have access to."
- "Last 6 hours to claim your March bonus"
- "Doors close tomorrow on the {Course} cohort"
- "Final reminder: invoice #4729 due in 24h"
Why it works: loss aversion is stronger than gain attraction. Real deadlines outperform manufactured urgency 3:1 — be honest.
5. Controversial take#
Stake a position your reader has an opinion on.
- "Marketing automation is mostly a waste of time"
- "Why you should send fewer emails, not more"
- "The Mailchimp pricing model is fundamentally broken"
Why it works: opinions invite engagement. Sycophants and cowards are forgettable. Have a real take; back it up in the body.
6. Direct question#
Ask the question they're already asking themselves.
- "Why is your open rate going down?"
- "What do you do when SES throttles you?"
- "Is your DMARC actually working?"
Why it works: if the recipient is asking this question themselves, you've met them where they are. Match audience pain — don't ask questions the reader isn't already thinking.
7. Two-step list ("3 things…")#
The mind expects "3 things" or "5 ways" — comforting structure.
- "3 emails I always send before a launch"
- "5 reasons your bounce rate is climbing"
- "7 templates that actually convert"
Why it works: numbered lists promise a clear, finite read. Stay honest — if your body has only 2 of the 5, the reader feels cheated.
8. Contrarian framing#
Take the opposite of the obvious framing.
- "Stop optimizing your subject lines (do this instead)"
- "Why I switched from Mailchimp to AcelleMail and regretted it for 3 weeks"
- "The argument FOR sending plain-text emails in 2026"
Why it works: the brain notices unexpected framings. Combined with curiosity (#1), this is high-leverage.
9. Time-of-day / day-of-week prefix#
Set the context up front.
- "[Tuesday] Your weekly digest"
- "[3pm] Quick win for the afternoon"
- "[Friday wrap-up] What happened this week"
Why it works: the prefix signals "this is timely, not generic" — slight bump in open. Also useful for newsletter-style cadence where consistency matters more than novelty.
10. Selective emoji#
One emoji, used sparingly, can lift open rate ~5-10%. Two or more = spam signal in many filters.
- "🎉 Your AcelleMail trial is live"
- "📊 Your March report"
- "🚨 Action needed: payment failed"
Why it works: emoji breaks the visual monotony of an inbox. Don't: use emoji in every send; use multiple emoji per subject; or use emoji that don't render in all clients (avoid newer ones like 🩷).
11. Plain-text feel ("from a person")#
Drop the marketing voice; sound like a human.
- "quick favor?"
- "wanted to share something"
- "did you see this?"
- "btw, about that thing you asked"
Why it works: marketing emails have a distinct voice; subverting it bypasses the "marketing filter" in the recipient's brain. Use sparingly — only when the body actually matches the casual tone.
12. The story tease#
Open with a fragment of a story.
- "I almost canceled the launch on Tuesday morning"
- "She told me my whole pricing model was wrong"
- "We sent 50,000 emails to the wrong list"
Why it works: stories activate different brain regions than marketing claims. Honest, specific, personal stories convert at 2-3× generic alternatives.
Worked example — combining formulas#
Subject lines often combine 2-3 of these. The Pro tier:
| Subject |
Formulas used |
| "{FIRST_NAME}, 3 reasons your open rate just dropped" |
#3 + #7 (personal + numbered list) |
| "🚨 Quick favor before tomorrow's deadline?" |
#4 + #10 + #6 (FOMO + emoji + question) |
| "Why we deleted 4,200 subscribers (and got 22% more revenue)" |
#1 + #2 (curiosity + specific number) |
| "[Tuesday] What broke our automation last week" |
#9 + #1 (time prefix + curiosity) |
Mix 2-3 deliberately. Don't try to cram all 12 into one subject line.
What NOT to do (spam triggers + open-rate killers)#
| Anti-pattern |
Why it fails |
ALL CAPS SHOUTING |
Triggers spam filters; reads as aggressive |
Excessive!!! punctuation??? |
Spam trigger |
| Generic words: "Free", "Save", "Discount", "Limited Time" |
Spam filters trained on these |
RE: or FW: to fake reply context |
Loss of trust; some filters detect + downgrade |
| > 50 characters on mobile |
Truncated in iOS Mail (40-50 chars visible) |
Long brand prefix: [ACME Newsletter] Weekly digest |
Wastes 18 chars before content; cut the prefix |
| Subject that doesn't match body |
Recipient unsubscribes; future opens drop |
| Same subject every send ("Weekly newsletter") |
"Already read it" → ignore |
The testing protocol#
You don't have to guess which formula works for YOUR audience. Test it.
Single-variable A/B test#
In AcelleMail (per A/B Testing guide):
- Pick TWO subject lines using different formulas (e.g. curiosity gap vs specific number)
- Send 50/50 to a 20% test segment
- Wait 4 hours
- Send the winning subject to the remaining 80%
Sample size: for stat sig, you need at least 300 sends per variant (with typical 20% open rate, this gives meaningful gap detection). On a 5,000-subscriber list, that's the 20% test segment.
Per-formula tracking#
Over 3 months, log:
| Send date |
Subject line |
Formula(s) used |
Open rate |
Click rate |
Patterns emerge. Some audiences love curiosity gaps; others respond to specific numbers; others to status/FOMO. Your audience is different from generic advice.
When to stop A/B testing#
- After 10+ tests with the same winner pattern, you've found your audience's preference
- After 5+ tests with mixed results, the difference doesn't matter for your audience — pick whichever you write fastest
Subject line length — quick reference#
| Length |
Where it cuts off |
When to use |
| ≤ 30 chars |
Even on Apple Watch |
Maximum mobile-friendly |
| 31-50 chars |
Visible in iOS Mail mobile |
Most marketing emails |
| 51-70 chars |
Truncated on most mobile clients; visible on desktop |
Only when first 30 chars are independently compelling |
| > 70 chars |
Cut on every mobile client |
Rarely worth it |
The first 30 characters do all the heavy lifting. If your hook is at char 45, you've lost.
Common failure modes#
| Failure |
Cause |
Fix |
| Open rate dropped 30% after good 6-month run |
Audience fatigue with your default formula |
Cycle through 4-5 formulas; don't repeat the same pattern weekly |
| Subject promised X, body delivered Y |
Marketer wrote subject for clicks, not honesty |
Always write subject AFTER body, not before |
| Emoji shows as 🔲 in some clients |
Used a newer/uncommon emoji |
Test in Apple Mail, Gmail Android, Outlook before sending |
| First-name personalization fails for some |
Missing FIRST_NAME field |
Use `{FIRST_NAME |
| Spam folder placement spike after good open rates |
Subject became too "marketing-y" |
A/B test plain-text-feel subjects against the marketing one |
FAQ#
Should I write subject lines first or last? Last. Write the email body first; let the strongest line from the body inspire the subject. Subject-first leads to subjects the body can't deliver on.
Does the subject line affect deliverability? Yes — content-spam-scoring algorithms (SpamAssassin, etc.) include subject line analysis. Triggers like ALL CAPS or excessive punctuation can move you to spam folder regardless of authentication.
How important is the "From" name vs subject? From name = ~50% of open decision; subject = ~50%. Investing only in subject without thinking about From name leaves opens on the table. See Email Marketing for Small Business... wait, this doesn't exist. Use a recognizable, consistent From name — usually the founder's name or your brand name, not "newsletter" or "no-reply".
Plain text emails — do subjects matter less? They matter MORE — plain text has no visual hook to compensate for a weak subject.
Should I A/B test From name + subject together? Test one variable at a time. Otherwise you can't tell which change drove the result.
Localization for non-English subjects? Same formulas apply; the cultural emphasis differs. Vietnamese: status/FOMO patterns are more powerful; Japanese: direct questions feel rude — soften with politeness. Test in your locale.
Related articles#
