Email subject lines and open rate: what works, what misleads, how to test

What a subject line is actually for

A subject line, together with the sender name, is usually all a recipient reads before deciding whether to open. It has one job: to help a busy person decide quickly whether the message is worth their time. That is a filtering task, not a creative-writing contest, and the subject lines that work tend to answer the silent question “what do I get if I open this?”.

A few principles hold across industries:

• Clarity: the reader knows what the email is about before opening it.
• Specificity: “Free delivery on orders over €50 until Sunday” beats “Special offer”.
• Relevance: the line suits the segment receiving it, not the whole database.
• Honesty: the promise in the subject is kept in the email. Misleading lines earn a few opens today and unsubscribes or spam complaints tomorrow.
• Measured curiosity: a little intrigue helps, but if the line hides everything, there is no reason to believe the email is worth opening.

The subject does not work alone. Who sends the message, when it arrives and what history you have with the reader all matter. A brilliant line sent to people who barely know you cannot make up for missing trust, which is why building your list properly usually moves results more than any headline formula.

How long should a subject line be?

There is no magic number. Email apps show different amounts of text depending on device, screen width and user settings. As a rough guide, a phone often displays somewhere around 30 to 50 characters of the subject, sometimes fewer, and a desktop client shows more. Longer lines are not forbidden, but the important part has to come first.

Practical rules:

• Put the key information in the first three to five words: the discount, the deadline, the benefit.
• Send yourself a test and read it on a real phone.
• Do not park essential details at the end, where they may be cut off.
• If you need more room, move part of the message into the preheader.

Our free email subject line tester shows how your subject and preheader may look in a few typical inboxes and flags risky wording before you send.

Preheader text: the second half of your subject line

The preheader is the short snippet shown beside or beneath the subject in the inbox. If you do not set one, the app pulls the first words of the email, which are often useless: “View this email in your browser” or navigation links. A deliberate preheader extends the subject instead of repeating it.

A simple model: the subject makes the promise, the preheader adds the detail. For example, subject “Three mistakes in e-invoicing”, preheader “And how to avoid them in under ten minutes”. Keep it to roughly 80 to 100 characters and check how it truncates in the apps your readers use.

Do not pad the preheader with invisible characters just to push out default text. It can look odd, and in some inboxes it may attract the attention of filters.

Weak subject lines and stronger rewrites

The examples below are hypothetical. They are not real campaigns and they do not promise any specific result; they simply show the difference in approach.

Look at the pattern. The stronger versions contain a concrete noun, a date or a number instead of a superlative. Not every line needs to be witty either. For a transactional message or a notification, the best subject is the most predictable one.

Spam filters: what really matters

Many people believe certain words, such as “free” or “discount”, send a message straight to spam. In practice, modern filters weigh many signals, and the words in the subject are only one of them. These usually matter more:

• Domain authentication: SPF, DKIM and DMARC set up correctly. Gmail and Yahoo introduced stricter requirements for bulk senders in 2024 (more than 5,000 messages a day): SPF, DKIM and DMARC authentication, one-click unsubscribe and a spam-complaint rate under 0.3%. Microsoft Outlook followed with similar rules in 2025, and non-compliant mail can now be rejected outright rather than only filed as spam, so check the providers’ current rules.
• Sender reputation: how many recipients open, click, report spam or unsubscribe.
• List quality: invalid addresses, purchased lists and subscribers who have been inactive for years damage deliverability.
• Content and formatting: image-only emails, links to dubious domains, broken HTML.
• A visible unsubscribe link: it reduces complaints.

For the subject itself, a few common-sense rules remain: text in all capitals, strings of exclamation marks, impossible promises, fake prefixes such as “Re:” or “Fwd:”, and wording that hides the commercial nature of the message.

Informational note: commercial email is regulated (consent, information duties, the right to unsubscribe). This text is not legal advice. Check your situation with a lawyer or your data protection officer.

What Apple Mail Privacy Protection changed

Since 2021 and iOS 15, Apple Mail Privacy Protection can preload the images in an email, including the invisible tracking pixel most platforms use to count opens. As a result, for some Apple Mail users a message shows up as “opened” even if nobody read it, and details such as the time or location of the open are no longer reliable. How much this affects you depends on how many of your subscribers use Apple Mail, which your email platform can usually tell you.

In practice this means:

• Open rates have been artificially lifted on many lists and are no longer a clean metric.
• An A/B test decided by “most opens” can crown the wrong winner.
• Automations triggered by an open (“if opened, send message two”) may fire for people who never read anything.

The answer is not to abandon open rate but to treat it as a trend line and lean more on clicks, replies, conversions and revenue per email sent. The same thinking applies to the flows described in our guide to essential email automations.

How to test subject lines, step by step

1. Write a hypothesis. For example: “a line with the deadline will earn more clicks than one with the benefit”.
2. Change one element. Only the subject, not the send time, sender name and content at once.
3. Split the list at random. Comparable groups of sufficient size. If your list is small, treat the results as indicative.
4. Choose the right metric. Clicks or conversions, not just opens.
5. Allow enough time. A few hours do not capture every subscriber; let it run for a day or two, depending on your list.
6. Record and reuse. A log of subject lines, hypotheses and results teaches you more than any list of “proven formulas”.

If your list has only a few hundred subscribers, a classic A/B test will not produce reliable differences. Compare results across several sends instead and learn from the trend rather than from small gaps.

Personalisation, emoji and capital letters

A first name in the subject can lift attention, but it carries risks: names in capitals or lower case, missing names, nicknames that do not suit the tone. Set a fallback (such as “Hello!” when the first name is blank) and check the data before sending. Behaviour-based personalisation, such as referring to an item left in a basket, usually works better than just inserting a name.

Emoji can help a message stand out in a crowded inbox, but overuse looks like spam and rendering varies between apps. One, placed with purpose, is normally enough, and for a B2B audience it is often safer to skip them. If you write in a language with accents, keep them in the subject: current apps display them correctly, and leaving them out looks careless.

Common mistakes

• Clickbait. A line that promises more than the email delivers loses trust over time.
• The same line every time. “Newsletter no. 14” gives nobody a reason to open.
• False urgency. “Last day!” every week teaches readers to ignore you.
• Judging by opens. Since Apple’s change, opens alone can mislead.
• Great subject, weak email. If the message does not deliver the promise, you won the open and lost the subscriber.
• Sending to the whole list. Basic segmentation (customers versus new subscribers) often changes results more than wording.
• Never checking on a phone. What looks fine on a monitor may be cut off in a pocket.

Key terms

Open rate: the percentage of recipients for whom an open was recorded, usually via a tracking pixel. Click rate: the percentage who clicked a link. Preheader: the preview text shown next to the subject. Deliverability: the likelihood that a message lands in the inbox rather than spam or promotions. SPF, DKIM, DMARC: methods of authenticating the sender’s domain. A/B test: comparing two versions on random groups. More terms like these are in our marketing glossary.

How we handle subject lines

When we take over email campaigns, we check the foundations first: domain authentication, list hygiene and segments. Only then do we work on subject lines. For each send we draft two or three options tied to the goal (click, booking, sale), test them where the list size allows and keep a record of results. We judge by clicks and conversions, not by opens alone, and we never promise a particular open rate, since it depends on the list, the offer and the season.

If you want help with the plan, the technical setup and the automations, see our page on email marketing and automation.

In short

• Write clear, specific, honest subject lines with the key words first.
• Use the preheader as the second sentence, not as filler.
• Spam placement depends more on authentication, reputation and list quality than on any one word.
• Since Apple’s privacy change, judge tests by clicks and conversions.
• Test one element at a time, keep a log and check everything on a phone.