The subject line is the most obsessed-over element in email marketing and one of the most misunderstood.
Add an emoji but don't add an emoji. Keep it under 50 characters. Use the word "you." Never use the word "free." Use curiosity gaps. Use numbers. Use power words - STOP PLEASE.
This advice has been recycled so many times that it has lost all meaning and most of it was based on open rate data, which is now so unreliable as a measurement that building a subject line strategy on it is roughly equivalent to making business decisions based on astrology.
So let's talk about what a subject line actually is, what it is actually supposed to do, and how to write one that works.
Most subject line advice is written as though the goal is to trick someone into opening. Make it so intriguing, so compelling, so irresistible that they cannot help themselves.
That is the wrong goal.
The subject line is a pre-promise. It is the first part of a two-part contract: here is what this email is about, and if you open it, here is what you will find. The preheader is the second part of that contract — it either builds on the promise or wastes the space entirely.
Together, the sender name, subject line, and preheader form the complete pre-open package. They are what the subscriber sees before they decide. And the decision they are making is not "is this exciting?" It is "is this for me right now?"
Those are fundamentally different questions — and writing for the second one produces very different subject lines from writing for the first.
Before someone reads your subject line, they read your sender name. And that sender name carries more weight than most email teams ever give it.
The sender name activates whatever association has been built with your brand in the inbox. If that association is positive — if past experience has coded you as useful, relevant, trustworthy — the subject line starts from a position of goodwill. The subscriber leans in slightly before reading a word.
If that association is neutral or negative — if you have been coded as noise, as promotional, as the sender they keep meaning to unsubscribe from — even a brilliant subject line is fighting uphill. The brain has already made its prediction before the subject line has had a chance to work.
This is predictive coding. And it is why the sender name is the most important thing in the pre-open package that nobody talks about when they talk about subject lines.
Practical implications:
Be consistent. Use the same sender name every time. Inconsistency breaks pattern recognition and feels suspicious. You want to be immediately recognisable.
Make it human where possible. "Beth at Astral" outperforms "Astral" in almost every context. A name signals a person, not a broadcast system.
Match the tone of your brand. A formal B2B sender name and a casual subject line create cognitive friction before the email is even opened.
Here is something the open-rate-maximising crowd does not want to admit: a high open rate from a misleading or clickbaity subject line is not a win. It is a set-up for failure.
When someone opens an email because the subject line promised something the email does not deliver — immediately, clearly, in the first few lines — two things happen. They close it. And they trust you slightly less than they did before.
Do that repeatedly and you have trained your audience that your subject lines lie. They start opening with scepticism rather than interest. Or they stop opening at all.
The approach I take — and the data consistently supports this — is clarity over cleverness.
Your open rate is not your success metric. Your click-through rate, your meaningful actions, and the quality of the people engaging are your success metrics. Write subject lines that attract the right opens, not the most opens.
This will annoy some people. Good.
Subject line A/B testing is one of the most widely practised and least useful things in email marketing. Here is why.
First: opens are an unreliable metric. Apple Mail Privacy Protection pre-loads emails, inflating open rates for any Apple Mail user. Outlook's reading pane suppresses opens. Corporate bots fire tracking pixels. The number you are testing against is not a clean measurement of human behaviour. You are optimising for a number that does not accurately reflect what you think it does.
Second: the sample sizes required for statistical significance in subject line testing are much larger than most email lists. If you have 5,000 subscribers and split them 50/50, you are testing with 2,500 people each. The margin of error on any result from a list that size means the "winner" is often within the noise. You are making decisions based on data that cannot support the conclusion you are drawing from it.
Third: even if you get a statistically valid result, it tells you almost nothing generalisable. Version A beat Version B with this list, at this send time, for this specific email type, in this cultural moment. That does not mean Version A will beat Version B next time — because the conditions are different.
What is worth testing instead? The things that are harder to test but more meaningful when you find an answer: the sequence of emails in a journey, the timing of a specific trigger relative to a behaviour, the presence or absence of a specific piece of content. Test at the programme level, not the subject line level. I've written about this in depth — the link is in the further reading below.
None of this means subject lines do not matter. They do. They are the pre-promise. They set the expectation. They determine who opens and in what frame of mind. Here are the psychological mechanisms that genuinely influence that decision.
The Zeigarnik effect is the psychological tendency to remember and be drawn toward incomplete things. A subject line that opens a loop — creates a question the brain wants to close — genuinely increases the likelihood of an open. Not through trickery, but through the basic mechanism of curiosity.
The distinction is important: an open loop should be a genuine loop. The email must close it. "How Linda grew her hair 10 inches in three months" works because the email actually tells you how Linda did it. The curiosity is real, the payoff is real. A subject line that opens a loop and then delivers something unrelated — or nothing at all — breaks trust.
Use open loops when you genuinely have something interesting to deliver. Do not manufacture mystery around ordinary content.
One of the most reliable subject line approaches is using the exact language your audience uses to describe their own problems. Not marketing language. Not the phrasing you chose when you positioned the product. The words people actually say out loud.
If your audience asks "why are my emails going to spam?" — that phrase in a subject line will outperform "best practices for inbox placement" every single time. Because it sounds like something they said, something they searched for, something that is about them rather than about you.
This is where audience research pays off directly in subject line writing. If you know the exact questions your audience asks, the exact complaints they have, the exact language they use to describe the problem you solve — you have unlimited subject line material that will consistently feel relevant to the people you are trying to reach.
"Grow your email list" is a vague claim. "How to add 200 engaged subscribers a month without paid ads" is specific. Specificity signals credibility. It implies the email contains something real rather than something generic. And it pre-selects: people who want that specific thing will open; people who do not, will not.
That pre-selection is desirable. You want the right people opening, not everyone.
A question in a subject line works when it is a question the subscriber has genuinely thought or said. Not a rhetorical device you invented — a real question from their real experience.
"Is your email list actually working?" lands differently for someone who has been quietly wondering exactly that than it does for someone who feels confident about their programme. For the first person, it is a direct address of something on their mind. For the second person, it is irrelevant. Both of those outcomes are fine. The first person is your audience.
Not specific examples — the approaches. The thinking behind them.
There are some basic technical realities worth respecting, even if they are not the main event.
Character length and mobile rendering. If your subject line is cut off on a mobile screen, the most important part of your message should not be at the end. On most mobile devices, you have roughly 30-40 characters before the cut-off. Know where your subject line ends and whether what the subscriber sees still makes sense.
The preheader is not optional. The space after the subject line in the inbox preview is one of the most wasted opportunities in email. It is where you extend the pre-promise, add context, or give one more reason to open. If you leave it empty, most email clients will pull the first text from the email — which is often a navigation link or an unsubscribe notice. Use the preheader deliberately.
Subject and preheader work as a unit. Write them together, not separately. The subject line opens the thought. The preheader completes it. If they say the same thing twice, that is wasted space. If they contradict each other, that is confusion.
Do not front-load with the sender name. Some ESPs automatically prepend the sender name to the subject line in certain clients, which can make your subject line start with "Astral: how to..." — effectively repeating the sender field. Know what your emails look like in real inboxes before you finalise anything.