How to Write Email Subject Lines That Get Opened

Nearly half of recipients, commonly cited around 47%, decide whether to open an email based on the subject line alone, before they see anything else. Get it wrong and the rest of the email, however good, never gets read. Get it badly wrong and you risk something worse: subject lines are also the single biggest factor recipients cite when deciding to mark an email as spam.

Clarity beats cleverness, almost every time

A subject line's first job is to set an accurate expectation, not to be impressive. In B2B contexts especially, straightforward subject lines that state what's inside consistently outperform clever or vague ones. Curiosity gaps can work, but only when the reader can guess the category of what's inside, just not the specifics.

Length: write for the phone in someone's hand

The majority of email is opened on mobile, where subject lines truncate well before desktop. Lines in the 30 to 50 character range consistently display in full across devices without cutting off your key message. Anything longer risks losing the part that actually matters.

Personalisation works, but only the kind that feels earned

Adding a first name or a relevant detail can meaningfully lift opens, but only when it reflects something the recipient would recognise as genuinely about them, like a past purchase or a stated interest. Personalisation that feels bolted on, a first name dropped into an otherwise generic line, reads as exactly what it is.

TIP

What the data says about specific tactics

Numbers in a subject line tend to lift opens, likely because they signal something concrete and scannable rather than vague. Questions tend to outperform plain statements. Emojis can help in the right context, particularly for younger or more casual audiences, but overuse or irrelevant use pushes toward looking like spam rather than adding warmth. None of these are rules to apply mechanically. They're levers worth testing against your own list, because what works varies by audience.

The words that used to matter less than people think

The old advice to avoid words like 'free' or 'sale' to dodge spam filters is largely outdated. Modern spam filtering from Gmail, Outlook, and Yahoo evaluates content holistically using machine learning, weighing sender reputation and recipient engagement history far more heavily than individual keywords. A single flagged word rarely does much on its own. What still matters is the pattern: excessive punctuation, all-caps, fake 'Re:' or 'Fwd:' prefixes on a message with no prior thread, and subject lines that misrepresent what's actually inside.

The sender name matters almost as much as the subject line

Recipients decide whether to open based on who sent something almost as often as what the subject line says. Keeping a consistent, recognisable sender name builds familiarity over time, and changing it abruptly can cost you opens simply because subscribers don't recognise you in the split second they're scanning their inbox.

Building a testing habit

The single highest-leverage habit for subject lines isn't finding a magic formula. It's testing two variants against a small portion of your list before committing to the winner for the full send. Over time, this builds a real picture of what your specific audience responds to, rather than relying on generic best practices that may or may not hold for your list.

Frequently Asked Questions

How long should a subject line be?

Aim for 30 to 50 characters so it displays in full on mobile devices, where most email is opened.

Do words like 'free' still trigger spam filters?

Not meaningfully on their own. Modern spam filters weigh sender reputation and engagement history far more heavily than individual words. Formatting patterns like excessive punctuation or misleading prefixes matter more.

Does personalisation actually improve open rates?

Yes, when it reflects something genuinely relevant to the recipient. Personalisation that feels generic or bolted on has far less effect.

What's the best way to improve my subject lines over time?

Test two variants against a small portion of your list before sending the winner to everyone. This builds an accurate picture of what your specific audience responds to.