Lifecycle
Free newsletter subject line tester and scorer
By Charles Summers · Updated · Free, no signup
Short answer
This measures a subject line rather than judging it. It counts characters against the mobile and desktop truncation points, counts words, computes the uppercase ratio across letters only, counts exclamation marks and special characters, detects emoji by code point, finds merge tokens, and compares your preview text against the subject to see whether it extends the sentence or just repeats it. Every deduction is printed with the number behind it, so you can disagree with any individual rule and still use the measurements.
Use the newsletter subject line tester
What does this tool actually do?
This measures a subject line rather than judging it. It counts characters against the mobile and desktop truncation points, counts words, computes the uppercase ratio across letters only, counts exclamation marks and special characters, detects emoji by code point, finds merge tokens, and compares your preview text against the subject to see whether it extends the sentence or just repeats it.
It runs entirely in your browser. Nothing you type is sent to a server, no account is required, and there is no usage limit, because there is no cost per run to control.
The inbox row has three fields and you are optimising the second one
A subscriber scanning an inbox reads a row, not a subject line. That row contains the sender name, the subject, and a fragment of preview text, in that order, and the first of the three decides most outcomes. Recognition is faster than comprehension: the eye resolves "Ledgerline" before it parses anything you wrote, and if the name is unfamiliar the subject rarely gets read at all. This is why a consistent sender name outperforms a clever one, and why rotating between a brand name, a person's name and a product name across campaigns quietly costs you engagement that no subject line rewrite recovers.
The sender column is also the narrowest and truncates hardest, especially in phone mail clients where it is competing with a timestamp and an unread dot. A from-name like "Ledgerline Product Marketing Team" is doing the work of "Ledgerline" plus twenty-three characters of nothing. If you have a personal sender that subscribers know, use it, and keep it stable long enough for recognition to build.
The preview text is the third field and the most wasted. Left empty, mail clients scrape the first readable text out of the HTML, which is why so many campaigns show "View this email in your browser" or an image alt attribute in the inbox. Filled properly, it is a second line of subject: it should continue the sentence rather than restate it. This tool compares the two strings token by token and penalises overlap, because a preview that repeats the subject has spent your only free extra line saying something the reader already read.
Where subject lines actually get cut, and why the number is a range
There is no character limit. There is a container width, filled with a proportional font, at a size the user controls. On a phone held in portrait, a mail client typically shows something in the region of 35 characters of subject before it stops, and the exact point moves with device width, with the app, with orientation, and with the accessibility text size the person has chosen. Desktop clients showing a list view give you roughly 60, and a three-pane layout with a narrow list column gives considerably less.
That is why this tool reports two thresholds rather than one verdict, and why it treats them differently. Passing 35 characters is not a failure, it is a statement about what happens to the tail: everything past that point is a bonus on a phone. Passing 60 means the end of your subject is invisible almost everywhere. The practical rule that follows is not "be short", it is "be front-loaded". Put the specific noun, the number, or the news in the first five words, and let the qualifier fall off the end without damage.
Word count is a separate and better-behaved measure, because words are the unit the reader scans in. Three to nine words is the range where a subject reads as one glanceable phrase. Past about twelve it becomes a sentence, and sentences in an inbox get skipped rather than read. A subject can be under 60 characters and still be too many words, which is why both numbers appear in the output.
A final structural note that the character count will not catch: a subject beginning with "Re:" or "Fwd:" on a message that is neither is a deliberate imitation of a personal thread. It is measurably effective at getting opened once, and it is one of the fastest ways to convert a subscriber into a complaint, because the reader feels tricked at exactly the moment they realise what they are reading. This tool penalises it heavily for that reason rather than a deliverability one.
Spam-word lists are folklore, and this tool says so on purpose
This tool contains a spam-trigger list. It is included because those terms genuinely correlate with badly written promotional copy, and flagging them improves subject lines. It is not included because it predicts what happens at the mailbox provider, and treating it that way will send you looking for the wrong problem. Modern filtering is dominated by things a text box cannot see.
What providers actually weigh: whether the message is authenticated by SPF and DKIM and aligned under a published DMARC policy, the reputation history of your sending domain and IP, how recipients have treated your mail before (opens, replies, moves to folders, deletions without reading), your complaint rate against the 0.3 percent bulk sender ceiling, whether you honour unsubscribes quickly, bounce rates that suggest a stale list, and structural signals such as image-only messages, public link shorteners, and a visible from-address that does not match the sending domain. Content classification exists, but it sits downstream of all that and is tuned per recipient, which is why the same message reaches one Gmail inbox and not another.
The reason word lists persist is that they are the only part of deliverability you can test in a form field. Reputation cannot be checked by pasting a sentence, so advice gravitates to the part that can, and an entire genre of tooling now scores copy and calls the result a deliverability prediction. It is not one. Say the word "free" in a subject line from a well-authenticated domain with an engaged list and you will land in the inbox. Send a beautifully worded message from a cold domain to a purchased list and you will not.
So read the flags here as editorial. A match tells you the phrase belongs to a register of writing that subscribers associate with junk, and that alone is a good enough reason to change it. It does not tell you where the message will land, and this tool cannot tell you that, because the answer depends on your sending history and on the individual recipient rather than on your words.
You cannot test subject lines on open rate any more
Open tracking works by embedding a remote image and recording the request when a mail client loads it. Since 2021, Apple's Mail Privacy Protection has pre-fetched remote content on behalf of users who enabled it, through proxy servers, whether or not the person ever looked at the message. The result is an open event with no human behind it. Because a large share of consumer mail is read in Apple Mail, the contamination is substantial, and because it varies with how many of your subscribers use Apple devices, it is not even consistent between two of your own segments.
This breaks three things at once. It inflates the headline number, which is merely embarrassing. It breaks A/B tests on subject lines, which is expensive, because the winner may simply have been sent to a slightly more Apple-heavy random split. And it silently corrupts anything built on opens: re-engagement segments defined as "opened in the last 90 days", automation triggered by an open, lead scores with an open component. Those systems now treat proxy fetches as human interest, which means your engaged segment contains people who have never seen a message.
The replacements are unglamorous and they work. Click rate measured against delivered, not against opens, removes the contaminated denominator. Conversion per thousand delivered ties the send to the outcome you wanted. Reply rate, for the kinds of newsletters where a reply makes sense, is unfakeable. For subject line testing specifically, judge the variants on clicks per delivered from the same send, and accept that you are measuring the subject and the first screen of the email together, because in practice that is what produced the click anyway.
One consequence worth planning for: subject line testing needs more volume than it used to. Click rates are smaller numbers than open rates, so the difference between two variants takes longer to separate from noise. If your list is small enough that a click-based test never reaches significance, stop running the test and spend the effort on the sender name, the send-time consistency and the offer, all of which move engagement more than a subject line rewrite does.
Numbers worth knowing
| Metric | Typical | What it means |
|---|---|---|
| Mobile subject truncation | around 35 characters | A portrait phone in a typical mail client. Moves with device width, app, orientation and the user's accessibility text size, so treat it as where the tail stops being reliable rather than a hard cut. |
| Desktop list truncation | around 60 characters | A three-pane layout with a narrow message list gives you considerably less. Front-loading matters more than total length, because the end is the part that disappears. |
| Usable preview text | roughly 40 to 100 characters | Enough to extend the subject into a full thought. Leave it blank and the client scrapes your HTML, which is how "View this email in your browser" ends up in the inbox. |
| Newsletter open rate | reported 15% to 45% | The spread across industries is enormous and the metric itself is unreliable, because Apple Mail Privacy Protection pre-fetches images and logs opens nobody performed. Use clicks per delivered. |
| Emoji effect on opens | no consistent direction | Published tests disagree, which usually means the effect is smaller than list and audience differences. Worth testing on your own list, not worth adopting as a rule. |
Mistakes that quietly cost you results
- Treating a spam-word score as a deliverability prediction
- Filtering is dominated by authentication, domain reputation, complaint rate and how your recipients have behaved before, none of which a text box can see. Fix the word because it reads as junk to a human, then go and check your DMARC record, which is where the actual risk lives.
- Leaving preview text empty
- The mail client fills it for you from the first readable text in your HTML, which is usually a browser-view link or an image alt attribute. You have surrendered a free line of copy in the inbox row and replaced it with navigation furniture.
- Writing preview text that repeats the subject
- The reader has already read those words about a fifth of a second earlier, so the duplicate line adds nothing and costs you the only chance to extend the thought. Write the two as one sentence broken in half: the subject makes the claim, the preview qualifies it.
- A/B testing subject lines on open rate
- Proxy image fetching means the winning variant may just have landed on a slightly more Apple-heavy split. Judge variants on clicks per delivered from the same send, and accept that small lists may never reach a readable result on that metric at all.
- Changing the sender name campaign to campaign
- The from-name is read before the subject and drives recognition, which is most of the open decision. Rotating between a brand, a product and a person resets that recognition every send, and no amount of subject line polish compensates for it.
What does the output look like?
This is the exact output the tool produces from the example inputs. It is generated by the same code that runs when you click the button, so what you see here is what you get.
Frequently asked questions
How long should a subject line be?
Between roughly 30 and 50 characters, and three to nine words, but the more useful instruction is to front-load. A phone in portrait shows around 35 characters of subject and a desktop list view around 60, and both figures move with the client, the window width and the reader's text size setting. Anything past those points is a bonus rather than a loss, provided the specific noun, number or piece of news sits in the first five words.
Does using the word "free" send my email to spam?
No, not on its own, and believing it does will send you looking in the wrong place. Mailbox providers weight authentication, sending domain reputation, complaint rate, bounce rate and each individual recipient's history with you far above the vocabulary of a single line. The spam-term list in this tool is an editorial flag: those phrases belong to a register of writing readers associate with junk mail, which is reason enough to rewrite them without pretending it predicts inbox placement.
Should I use emoji in subject lines?
Published tests point in both directions, which usually means the effect is smaller than the differences between lists and audiences. This tool detects emoji by code point and reports them without a penalty unless there are several, because a row of symbols reads as promotional shouting. Two practical cautions: emoji render differently across platforms and occasionally as an empty box, and they are the single most common cause of an encoding surprise if the same copy is reused for SMS.
Why does the tool compare my preview text to my subject?
Because the two fields are read as one unit, roughly a fifth of a second apart, and duplication wastes the second one. The check normalises both strings, removes trivial words, and measures how much of the preview's vocabulary already appeared in the subject. High overlap, or a preview that literally begins with the subject, triggers the finding. The fix is to write them as a single sentence broken in half rather than as two attempts at the same sentence.
Can this tool tell me whether my email will land in the inbox?
No, and no tool that only reads your copy can. Inbox placement depends on your authentication records, the reputation your sending domain has accumulated, your complaint and bounce rates, list hygiene, and how the specific recipient has treated your previous messages. This tool measures readability and convention: how the line will render, how it will scan, and whether it follows the norms subscribers expect. Those are worth measuring, and they are not the same thing as deliverability.
Related free tools
- Push Notification Writer Lifecycle
- Onboarding Email Sequence Writer Lifecycle
- Win-Back Email Writer Lifecycle
- Referral Program Designer Sales
- Lead Scoring Model Builder Sales
Some links on this site are affiliate links, which means Hacking Demand may earn a commission if you buy through them at no extra cost to you. This does not influence which tools are listed. The tools on this page are free and have no affiliate relationship of any kind.