Improving open rates: what a subject line can do and what it cannot
The subject line gets all the attention and is the third most important factor. Two others come before it: whether the message reaches the inbox at all, and whether the sender is familiar.
In short
- Since images began being preloaded automatically, the open rate measures technology to a considerable extent rather than interest.
- Three things decide the actual opening, in this order: delivery, sender name, subject line.
- The sender name is the strongest lever and almost never gets changed — a person's name works differently from a company name.
- Clicks on a particular link, and replies, are the more dependable metrics.
Why the number says less than it used to
An open is measured through a tiny image loaded when the message is displayed. Since several large mailbox providers began preloading images through their own servers, that image gets fetched even when nobody reads the message.
The consequence: the open rate is systematically too high for recipients at those providers and too low for others — and the make-up of your recipient list determines how strong the effect is. Comparisons between two sends stay roughly usable; absolute values do not.
Worth knowing
The strongest lever for opens is not the subject line but the sender name — and in most companies it gets set once and never looked at again.
A person's name with a company suffix ("Emma Woelk – TEISENDA") is treated differently from a bare company name, and both differently from an impersonal address. The reason is recognition: in the inbox, recipients decide from the sender first whether to read the subject at all. That change takes two minutes and works more strongly than any subject line optimisation.
The four levers, ordered by effect
1. Delivery
What lands in spam does not get opened. SPF, DKIM and DMARC in place, a maintained list with no dead addresses, your own sending domain.
Effect: very high · Effort: high, once
2. Sender name
A person's name with a company suffix, kept the same across all sends. An address that can be replied to.
Effect: high · Effort: two minutes
3. Subject line
Name the outcome, do not tease. 30 to 50 characters, so it is fully visible on a phone.
Effect: medium · Effort: per send
4. Preview text
The line after the subject in the inbox. It usually gets left unused and then shows the start of the message — often "If this email does not display correctly".
Effect: medium · Effort: one minute per send
What a subject line can do
Four rules that hold up in practice:
- Name the outcome. "Why your mail lands in spam — and what helps in each case" instead of "March newsletter".
- Short enough for a phone. 30 to 50 characters. What gets cut off does not exist.
- No manufactured curiosity. "We did not see this coming …" gets opened once and ignored afterwards. The short-term gain costs trust in the long run.
- No urgency without a reason. "Today only" in an automated sequence is visibly untrue, because everyone receives it at a different moment.
The preview text is the most frequently wasted space in the whole inbox. If it is not set, many programs show the start of the message — and in many templates that is a technical note or an unsubscribe line.
It works as a continuation of the subject, not as a repetition: the subject names the topic, the preview text the specific figure or outcome. One minute of effort per send for the second largest visible space.
What to measure instead
| Metric | What it shows | Distorted? |
|---|---|---|
| Open rate | technology plus some interest | heavily |
| Clicks on a particular link | actual interest | barely |
| Replies | genuine willingness to talk | no |
| Unsubscribes | whether the expectation was right | no |
| Delivery rate and bounces | technical health | no |
The second row is the most useful for day-to-day running: clicks on a clearly named link show whether the topic landed — and they can be compared between sends.
Help me phrase a subject line and preview text for a send. Content of the message: [paste text or summary] Details: - Recipient group: [description] - What recipients get out of it: [specific outcome] - Sender name we use: [details] - The one link that should be clicked: [description] Tasks: 1. Assess our sender name: will it be recognised in the inbox? Propose a better form if appropriate. 2. Phrase five subject lines that name the outcome – 30 to 50 characters, no manufactured curiosity, no urgency, no first name. 3. Phrase a matching preview text for each subject line that continues the subject rather than repeating it – with the specific figure or the specific outcome. 4. For each version, tell me what would be cut off on a phone. 5. Name which metric I should look at for this send instead of the open rate. Use only details from the content above.
In closing
The subject line matters and comes third. Before it, delivery and sender name decide — and both get set up once rather than reinvented at every send.
And the number itself is barely usable any more: clicks on a named link and replies say more about the effect than any open rate.
Common questions
Why is the open rate no longer reliable?
Because it is measured through an image loaded on display — and several large mailbox providers preload images through their own servers even when nobody reads the message. The values are therefore too high for some recipients and too low for others; rough comparisons between two sends stay usable, absolute values do not.
What influences opening most strongly?
Three things in this order: whether the message gets delivered at all, the sender name, and only then the subject line. The sender name is the most underrated lever — in the inbox, recipients decide from the sender first whether to read the subject at all.
How long should a subject line be?
30 to 50 characters, so it is fully visible on a phone. What gets cut off does not exist for the recipient — and the most important part often sits exactly there.
Does a first name in the subject help?
Not any more. The device is so common in promotional mail that it reads more like a mass send than like personalisation. Personalisation works only when it refers to something the person actually did.
What should you measure instead of the open rate?
Clicks on a clearly named link — that shows actual interest and can be compared between sends. Alongside it: replies as a sign of genuine willingness to talk, unsubscribes as an indication of wrong expectations, and delivery rate together with bounces as a technical health check.
Marketing that sets itself up
The Studio Engine beta is live. Claim your spot and help shape it from the start.
Join the beta →