Free · no signup

Cold email subject lines that get opened.

Plug in your offer, the role you're emailing, and their company. Get B2B subject lines across 7 proven formulas — curiosity, pain-point, outcome, pattern interrupt — in your tone.

TL;DR

Cold-email open rates live and die on the subject line and the sender — not the body. The rules that actually move the needle: keep it under ~50 characters (most B2B email is previewed on mobile), make it specific to the reader, and lead with curiosity, not a pitch. Subject lines that read like a 1:1 note from a colleague beat polished "campaign" subjects almost every time.

This tool fills seven proven cold-email formulas with your real inputs, so the personalization is structural rather than a {{merge-tag}} swap. Generate a batch, pick the 3 strongest angles, and let reply rate — not open rate — decide the winner. (Deliverability still depends on domain warm-up, volume, and list quality — a great subject can't rescue a cold domain.)

Cold Email Subject Lines → Division50

Great subject lines are 5% of outbound.

The other 95% is list quality, multi-channel cadence, deliverability, and a human who follows up. Division50 runs the whole outbound engine — real SDRs, calls + email + LinkedIn — and books qualified meetings on your calendar.

Book a strategy call →

30-minute call · no deck, no obligation · global remote operators.

The 7 formulas — and when to use each

Curiosity & questions open a loop the reader wants closed — best when your specific question is genuinely relevant. Personalized references something true about them, not a merge tag. Pain-point names a problem they already feel without overstating it. Outcome leads with the result and drops the mechanism. Pattern interrupt is short, human, and unexpected — it stops the scroll in a busy inbox. Social proof borrows credibility (keep the numbers honest). Direct works when the offer is strong enough to say plainly.

How to use the output. Don't ship all of them — generate the batch, then pick the 3 strongest angles (not just wordings) and A/B test those, one variable at a time. Measure reply rate, not open rate: opens are increasingly inflated by privacy proxies, and a meeting booked is the only metric that pays. Pair the winning subject with a tight, single-CTA first line — the subject earns the open, the first line earns the read.

This generator is a starting point for ideation. It doesn't send email or store anything you type — everything runs in your browser.

Frequently asked questions

What makes a good cold email subject line?
Short (under ~50 characters so it doesn't truncate on mobile), specific to the reader, and curiosity-driven — not a pitch. The best B2B subject lines read like a 1:1 note from a colleague, not a campaign. Lowercase and lightly imperfect often out-performs polished, because it feels personal. Never put your offer in the subject if you want it opened — save that for the first line.
How long should a cold email subject line be?
Aim for 1–5 words / under 50 characters. Most inboxes truncate subjects around 40–60 characters on desktop and far less on mobile, where the majority of B2B email is now previewed. Shorter subjects also read as more personal — a 2-word subject signals 'a human wrote this for me,' a 9-word subject signals 'mass send.'
Should I use personalization in the subject line?
Personalize the relevance, not just the merge tag. Dropping {{company}} into every subject is obvious and ignored. Better: reference something true and specific — a role they're hiring for, a market they're in, a problem their function feels. This tool fills proven formulas with your real inputs so the personalization is structural, not a token swap.
What words get cold emails sent to spam?
Overtly salesy or hype words (free, guarantee, act now, limited time, $$$), ALL CAPS, excessive punctuation (!!!), and too many links. Deliverability is mostly about domain reputation, warm-up, and volume — but spammy subjects compound the problem. Keep it plain and human.
How many subject lines should I A/B test?
Test 2–4 at a time, one variable at a time, with enough volume per variant to reach significance (a few hundred sends minimum for a directional read). Test the angle (question vs outcome vs pain) before micro-optimizing wording. Generate a batch here, pick the 3 strongest angles, and let reply rate — not open rate — decide the winner.