Cold Email Subject Lines: What Actually Gets Opened

A subject line has one job: get the message opened without triggering a spam filter or a reader's skepticism before they even see the body. Most subject-line advice optimizes for the first half and ignores the second, which is why "clever" lines increasingly underperform plain ones.

What consistently works

Specificity beats cleverness. "Question about [specific process at their company]" outperforms "Quick question!!" because it signals the sender actually knows something about the recipient, and it gives the filter nothing to flag.

Short enough to survive mobile truncation. Gmail's mobile app shows roughly 30–40 characters of subject line before cutting it off. A subject line whose meaning depends on its second half is read as a fragment by most recipients, who open on a phone. Front-load the useful part.

Lowercase and sentence case, not Title Case. Subject lines written the way a person types — lowercase, no unnecessary capitalization — read as personal rather than templated. "quick question about your onboarding flow" reads differently than "Quick Question About Your Onboarding Flow".

A name or specific detail, when it is genuinely true. "re: [specific project or event]" works when there is a real connection to reference, and reads as deceptive the moment there is not.

Consistency with the body. The subject line should be answerable from the first sentence of the email. A subject that promises one thing and a body that pivots to another is the fastest way to earn a spam report on open.

What reliably hurts deliverability or opens

• ALL CAPS, excessive punctuation ("!!!"), and classic trigger phrases ("free", "guarantee", "act now", "no obligation") are weighted by Bayesian spam filters and by human pattern recognition alike. Age has not made these safer; filters have only gotten better at catching them.
• "Re:" or "Fwd:" on a message that is not actually a reply or forward. This is deceptive-header territory, flagged by increasingly sophisticated filters and, where it violates anti-spam law, a compliance risk, not just a deliverability one.
• Emoji in B2B cold outreach. They work in some consumer and creator contexts; in professional cold outreach they read as unserious and occasionally trip filters tuned for promotional mail.
• Vague curiosity-bait ("you won't believe this") that B2B recipients have learned to recognize as low-value mass mail. It occasionally works in consumer contexts; in B2B it signals template, not person.
• A subject line that does not match the sender. If the from-name is a real person but the subject reads like a company-wide announcement, the mismatch itself is a soft signal something is automated.

A short checklist

Testing without guessing

Subject-line "best practices" generalize badly because performance depends heavily on your specific list, industry and existing sender reputation — a subject line that performs well for one audience can underperform for another with no obvious reason. Test on a small batch before committing a full send, and track opens against the same seed-inbox placement data you use for the rest of the campaign, because a subject line cannot be judged fairly if some of the test batch never reached the inbox at all.

A domain with weak sender reputation will see a flat, low open rate no matter how good the subject line is — the message is landing in spam before anyone reads the subject. If open rates are uniformly low across every variant you test, the problem is probably the domain, not the copy.

Examples: the same message, three ways

A B2B outreach email offering a free audit of a prospect's onboarding flow, written three ways:

Weak (trigger words, vague):

Free Audit - Act Now, Limited Spots!!!

Better, but still has a problem:

Re: your onboarding flow - noticed something

This one is specific, but the fake "Re:" on a message that isn't actually a reply is deceptive-header territory, and it front-loads a vague hook rather than the concrete detail.

Works well:

quick question about [Product]'s signup flow

Short, lowercase, names something real and specific, and the body can open with exactly that question. All of it displays before mobile truncation, and the meaning is already clear within the first 30 characters.

Why mobile truncation changes the calculus

Most B2B cold email is now opened on a phone first, even when the recipient reads at a desk later — notification previews and quick triage happen on mobile. A subject line that reads naturally at 60 characters on a desktop inbox can arrive as a meaningless fragment on a phone lock screen. Write for the 30-40 character window as the primary constraint, and treat anything beyond that as a bonus for desktop readers rather than content the message depends on.

MailPilot's per-provider inbox placement tracking is how you tell a copy problem apart from a domain problem before you spend another week rewriting subject lines that were never the issue.