Email Subject Lines: What Actually Decides the Open

The results for this term are mostly lists of 150 subject lines you can copy.

They are worse than useless. A subject line that worked on someone else's list is evidence about their relationship with their readers — their brand, their sending history, their offer. It is not evidence about words.

Paste it into your own send and you keep the words and lose everything that made them work.

If you are still setting the channel up, start with our guide to email marketing and come back. This page assumes you already send and want the opens to be real. For the research habit behind any good line — reading how customers describe the problem in their own words — we keep that in copywriting and will not repeat it here.

The three things a reader sees before deciding

Every inbox row shows the same three fields: who it is from, the subject, and a slice of preview text. The reader scans all three in one glance, which is why treating the subject line as a standalone craft problem is the core mistake.

Above all three sits history. If the last five emails from that sender name were worth opening, a mediocre subject still gets opened. If they were not, a brilliant one gets archived unread.

That is why the listicles fail, and why subject lines are usually the third thing to fix, not the first.

What the subject line's job actually is

Its job is to earn the open without over-promising the body. The second half is the expensive one.

An over-promise is paid back in unsubscribes and complaints, and both are priced into whether your next email arrives at all. Gmail Help's guidance for bulk senders (no last-updated date shown on the page; read 1 October 2026) tells senders to "keep your spam rate below 0.1%" and to "prevent spam rates from ever reaching 0.3% or higher", and says that from June 2024 senders above 0.3% "will be ineligible for mitigation".

Leaving is also easier than it was. Google's subscription guidelines require "one-click unsubscribe for all subscription messages, in a way that no additional user action is needed to unsubscribe", and that senders "process and honor unsubscribe requests within 48 hours".

So a clickbait subject buys one open and can cost you the address. Google's bulk email best practices page is blunt: "use clear subjects and headlines to clearly communicate your goals", and deception "is directly against Gmail's policy".

The sending side — authentication, complaint rates, list hygiene — sits in email deliverability.

Patterns that work, and the condition each one needs

Patterns are not templates. Each works only when a specific condition is true, and the condition is the part the listicles leave out.

Length and truncation: what nobody publishes

No major mail client publishes a subject-line character limit, and we will not invent one. Truncation is decided by pixel width: the device, the reader's font size, orientation, dark mode and whether a preview pane is open all move where the cut lands.

• The decision happens in the first few words. On a phone the visible part is roughly the first fragment; treat the rest as a bonus that may never render.
• Front-load the subject, not the brand. The brand is already in the sender name.

Instead of chasing a number, test what you can see: send the campaign to yourself and open it in the Gmail app on a mid-range Android, on an iPhone, and in desktop Outlook with the reading pane open. What gets cut there is cut for most of your list.

Long is not automatically worse. What loses is a subject whose meaning only arrives at the end.

Preview text is the second line of your headline

Preview text — the preheader — is the sentence after the subject, not a repeat of it. Write the two together and read them aloud as one line.

Leave it blank and the client fills it with whatever comes first in the email: "View this email in your browser", an unsubscribe line, the alt text of a logo. You have handed your second headline to a template.

1. Continue, don't repeat. The subject raises the specific thing; the preview text adds the qualifier, the date, or the catch.
2. Useful part first. It truncates like the subject, often harder.
3. Keep it honest to the body. It is the last thing checked before a reader decides you over-promise.
4. Check it is not empty. Audit a past send and see what actually showed.

Consumer email in Gmail has one extra wrinkle. Google's documentation for Promotions tab annotations (last updated 8 September 2026) describes a deal annotation that "adds a promotional badge next to the subject line", and notes that "a variety of factors affect whether email annotations are shown to users, including quality filters and frequency limits". So the row a retail reader scans may carry a badge you did not write.

Why open rate no longer tells you whether the subject worked

Open rate is measured by a tracking pixel: an image that loads when the email is displayed. Apple broke that signal deliberately.

Apple's Mail Privacy Protection documentation (apple.com/legal, page dated 12 December 2025) states that "rather than only downloading remote content when you open an email, Protect Mail Activity downloads remote content in the background by default", and that it "helps protect your privacy by preventing email senders, including Apple, from learning information about your Mail activity". The same page describes routing that content through two separate relays, so that no single party knows both the reader's IP address and the content received.

Read it as a measurement fact. For every subscriber using Apple Mail with this on, your platform logs an open whether or not a human looked. Your reported open rate mixes real opens with automatic fetches, in a ratio that shifts with your list's device mix.

The honest alternatives:

• Click-to-delivered. Clicks divided by delivered, not by opens — click-to-open rate inherits the inflated denominator.
• Reply rate. On small and mid-sized lists, a reply is the one signal nothing can fake.
• Unsubscribe and complaint rate per send. Your over-promise detector, priced by Google at the thresholds above.
• A variant compared against itself over time. Same segment, same send slot, same offer type, across several sends — never against a published benchmark, which averages other people's lists measured with the same broken pixel.

Testing subject lines properly on a small list

Most SMB subject-line A/B tests are not readable. The list is too small for the size of difference being tested, so the "winner" is noise and you learn a false lesson you then repeat.

Check it before you send. Using the standard two-proportion sample-size formula at 95% confidence and 80% power, from a 20% baseline:

Those are computed from the formula, not taken from a study — put your own baseline into our A/B test calculator for the same shape of answer. The method sits in our A/B testing guide.

If your segment is in the low thousands, do this instead:

1. Stop testing small differences. Test a different angle, not a different adjective.
2. Test the whole email on clicks. Subject, preview text and offer as one package, judged by click-to-delivered.
3. Pool across sends. Run one structural choice — specific-number subjects against question subjects — over six sends and compare the groups.
4. Keep a decision log. One line per send: what you tried, what you saw, what you will do next.

Expect a quarter of sends before a real pattern shows, not one campaign.

Rewrites: generic on the left, specific on the right

These are illustrations written for this page to show the move from vague to specific. They are not tested results, and no claim is made about how any of them performed.

Every rewrite is also a promise the body has to keep. Before sending, ask whether the first screen of the email pays it off.

Subject lines for Indian lists

Two audiences behave differently enough that one house style will not cover both.

• English-first professional lists (B2B, SaaS, education, finance) are often read on desktop, and the sender name does most of the work. Plain, specific subjects read as a person writing; a leading emoji reads as a bulk send.
• Consumer and retail lists are mobile-first and frequently land in Gmail's Promotions tab, where the row already competes with badges and images. An emoji can help the eye find the row, and the offer belongs in the first fragment.

No vendor publishes a penalty for emoji or ALL-CAPS and we will not quote one. Both are audience-specific, so they belong in your own pooled testing rather than a rule copied from a global blog. Script is the same decision: if your list genuinely reads Hindi or Marathi, the subject can, but only if the body and the landing page continue in that language.

How we write them (Digital Hangover's practice)

This is our working habit on client lists, offered as practice rather than proof — we are not attaching a figure to it, because we have not run a controlled test of the habit itself.

We write the preview text first. It forces the useful sentence out early, and the subject becomes the shortest possible pointer to it rather than a clever line we then have to justify.

Before touching words, we read the last five emails that went to that list, in order, as a subscriber would. Half the time the subject line is not the problem — the previous sends were not worth opening, and no wording fixes that. When that is the finding, we say so and the work moves to the offer and the cadence. That is the kind of call our content marketing services are scoped to make.

Where to go from here

Fix the three fields, not the one. This week: set a consistent sender name, write real preview text for every template that has none, and switch your reporting to click-to-delivered.

Then audit your last ten subject lines against one question — did the body keep the promise? Where it did not, fix the body. The same specificity rules apply in a tighter space if you write paid ads: see ad copy.