What is a good unsubscribe rate?#
As a starting figure, a rate under about 0.3 percent of delivered mail is ordinary for a list that people chose to join and that gets mailed on a steady schedule. Between roughly 0.3 and 0.5 percent is worth a look at who you mailed and why. Above that, on one send or on several in a row, something specific changed and the number is telling you where to look rather than telling you that your writing is bad.
That band is a working reading, not an industry standard, and it is worth saying so plainly because a lot of pages present a number like that as if somebody measured it. The measured numbers are the ones a vendor publishes from its own customer base. Everything above them is interpretation, and interpretation is where two tools that both have real data can disagree.
Before any of that is useful, get the definition fixed in your head, because the same percentage can describe three different actions. An unsubscribe is somebody clicking the link that takes them off your list. It is not somebody reporting you as spam, and it is not somebody who stopped opening. Only the first one is your own choice being honoured, and it is the cheapest of the three to absorb.
How is an unsubscribe rate calculated?#
MailerLite defines its figure as the percentage of recipients who click the unsubscribe link in your email. That is the plain reading, and it hides one decision that moves the number every time: the denominator. A rate can be built on everyone the campaign was addressed to, on everyone whose mail a receiving server accepted, or on everyone whose message was opened. The same eight unsubscribes produce three different percentages depending on which of those you divide by.
The reading that behaves best across two campaigns is unsubscribes divided by delivered. It excludes addresses that never received anything, so a dirty list cannot flatter your rate, and it does not depend on an image being fetched, so a privacy proxy cannot distort it. Whichever you use, use the same one every time and write it down, because a rate you cannot compare to last month's rate is decoration.
What numbers do email tools publish about unsubscribes?#
These are two vendors describing their own customer data, taken from their own pages. They are not a universal law, and the difference in how each was measured is the useful part.
| Who published it | Their unsubscribe figure | What it was measured across | What they say about the change |
|---|---|---|---|
| MailerLite, 2025 benchmarks | Median 0.22 percent | 3.6 million campaigns sent by its own customers | More than double its 2024 figure of 0.08 percent |
| MailerLite, by industry | Highest 0.40 percent, lowest 0.09 percent | The same dataset, split into 46 industries | Photo and video highest, legal industry lowest |
| Mailchimp, benchmarks resource | Average 0.22 percent for all users | Billions of emails delivered through its system, campaigns of at least 1,000 subscribers | 0.15 percent business and finance, 0.18 non-profits and education, 0.19 ecommerce, and its own page says the data was last updated in December 2023 |
Read the middle row slowly, because it deflates the whole idea of an industry benchmark. The spread across MailerLite's own industries, from 0.09 percent at legal to 0.40 percent at photo and video, is over four times. MailerLite puts it in people rather than percentages, which is the right way to hear it: at the top of that range, roughly three to four out of every thousand recipients unsubscribe. A benchmark averaged across forty-six industries describes nobody in particular.
There is also a reason the 2025 figure is higher than 2024, and MailerLite names it rather than hiding it: the changes Gmail made that let somebody unsubscribe from a promotional email in one tap in the inbox, without ever opening it, plus a manage subscriptions screen. Their own conclusion is that the rate went up because the action got easier, not because their customers started writing worse emails. A reader who does not know that reads 0.22 against 0.08 and concludes they have been sabotaged by a subject line.
Is a high unsubscribe rate always bad?#
No, and it is the distinction most advice on this query skips. An unsubscribe is a person using the exit you offered, which is the mechanism that keeps them from reporting you as spam instead. A list that nobody ever leaves is usually a list that nobody opens either, and the harm there is invisible in the report. MailerLite states the trade plainly on its own benchmarks page: it is better to have a smaller list that engages with your content than a larger one that never opens your emails.
There is a specific case where a spike is the result you planned. A re-permission email, sent to people who have gone quiet, exists to get a yes or a leave, and the people who leave are the answer you paid for. The same is true of the first campaign after you switch from mailing your whole list to mailing an engaged segment: the remainder, suddenly mailed something they did not ask for, unsubscribe, and the send after that gets quieter and better.
What is the difference between an unsubscribe and a complaint?#
An unsubscribe is addressed to you. A complaint is addressed to the mailbox provider, and it is the number that provider scores you on. Google tells senders to keep the spam rate it reports in Postmaster Tools below 0.3 percent, and Microsoft describes complaint rate as the principal factor in how it judges a sender. Neither of those public ceilings is about unsubscribes.
That difference is why a rate that looks polite can be the dangerous one. Ten unsubscribes and ten spam reports are the same count of people leaving, and the second group is costing you a reputation that takes weeks of clean sending to rebuild. If you can only watch two numbers on a campaign, watch unsubscribes as feedback about the content and complaints as feedback about the list.
Why did my unsubscribe rate suddenly jump?#
Work through the causes in order of how common they are. Most jumps have a boring explanation that has nothing to do with the writing.
| The shape of the jump | What usually explains it | What to check before changing anything |
|---|---|---|
| One campaign spikes, the next is normal | A subject line that promised something the email did not contain, or a topic outside the segment it went to | Read the subject against the first paragraph, and check who the segment matched |
| Every send climbs slowly for months | Cadence crept up, or the same addresses keep getting mailed after they stopped reading | Count sends per person per month, and look at how many contacts have not opened in two quarters |
| A jump right after an import | Names that never asked arriving from a spreadsheet, a file that came from elsewhere, or an old list re-added | Check whether the imported rows went out as a separate audience and whether confirmation is switched on |
| A jump on one segment only | A tag or rule that quietly changed meaning, so the campaign reached people it was not written for | Open the saved rule and count who it matches now, not who it matched when you saved it |
| A jump alongside a bounce spike | An old list being mailed at all, where the dead addresses are the same people who no longer want it | Sort the two counts apart before reading either rate |
| A jump with no complaints and no bounces | A seasonal or one-off audience, like attendees, customers from a single campaign, or a giveaway | Decide whether that audience wanted a relationship or an answer |
How do I find my own baseline instead of an industry one?#
- Write down the unsubscribe rate of your last six to twelve campaigns, using one denominator for all of them.
- Take the middle value rather than the average, because one re-permission send or one bad import will drag a mean into a number you then chase.
- Cross out any campaign that was not a normal send to a normal audience, and note why, so the exception is not quietly setting your standard.
- Compare the next three sends against that median rather than against a chart from somebody else's customer base.
- Move the baseline only when the audience or the cadence has changed, which is roughly once a year for most small lists.
A list of 1,000 people mailed twice a month at a 0.2 percent rate loses about two people per send, so about fifty a year. That is the arithmetic that makes a small percentage feel large, and it is also the reason a rate is better read as people than as a decimal. Fifty a year on a list that adds two hundred is churn you can live with. Fifty on a list that adds ten is the same percentage and a different business.
How do I lower my unsubscribe rate without tricks?#
Four changes move this number, and none of them are a cleverer subject line. They are all about whether the person on the other end got mailed something they had reason to want.
- Confirm the signup. MailerLite recommends double opt-in specifically as an unsubscribe control, and it is the one step that changes the composition of the list rather than the composition of the report.
- Mail segments rather than the whole audience, and let the rule be about behaviour you have observed rather than about when somebody joined.
- Hold a steady cadence and say it on the signup form, because irregular volume reads as spam to a reader who forgot asking.
- Re-permission the quiet half instead of mailing it indefinitely, which costs unsubscribes once and then produces a smaller list that behaves better.
Two moves that look like fixes are not. Making the unsubscribe link hard to find keeps the person, converts the unsubscribe into a complaint, and is illegal in the United States under the rules the FTC publishes, which require a clear way to opt out and prompt honouring of the request. Deleting the unhappy half of a list without ever offering them a choice produces a beautiful rate on an audience that no longer exists.
What happens to somebody who unsubscribes?#
They should stop receiving mail from that point, everywhere, and permanently, unless they ask to come back. In practice that means one record that survives the campaign they left on, so a second tool or a fresh import cannot mail them again from an old copy of the list. Yahoo's public requirement is that a one-click unsubscribe is honoured within two days at most, and a sender that takes a week has a broken process rather than a slow one.
In Campaign Kicker an unsubscribe is written as an event on the contact, not only on the campaign, and the address then drops out of every later send, automation and segment on that account. If you mail from two tools at once during a move, that is the moment to check both of them honour it, because a list copy is where unsubscribes go to be re-mailed.
Should I stop sending to people who unsubscribe and keep the rest?#
Yes, and the question is about whether to keep the record. Delete nothing. A suppressed address is the proof that you honoured a request, which is what a provider, a payment processor or a regulator asks for when somebody claims you mailed them anyway. Keep the address, the date and the campaign it came from, and hold it out of every send. That is the version of list hygiene that survives an argument.
Does a big unsubscribe number mean my list is too small?#
It usually means the opposite of what people fear: your list is small enough that one campaign is visible. On a list of 200 people, five unsubscribes is 2.5 percent, which looks catastrophic and is a rounding error. On a list of 20,000, five unsubscribes is 0.025 percent and might be the same five unhappy readers. Percentages on small audiences are noise, so read counts until the audience is big enough for the rate to mean anything.
Do unsubscribe rates differ by industry?#
Yes, and the published spread is wider than most advice online assumes. MailerLite breaks its own 2025 dataset into 46 industries and puts photo and video at the top with 0.40 percent, followed by restaurants and cafes at 0.39 percent and telecommunications at 0.34 percent. At the other end it reports the legal industry at 0.09 percent, with media and higher education both at 0.1 percent.
The honest reading of that table is not that lawyers write better emails. A legal newsletter goes to people who need it as a working reference and who leave when they stop needing it, which is a slow, orderly exit. A photo competition mailing goes to people who joined for one particular thing. Same action, different reasons, and a single averaged benchmark hides both. The regional cut in the same article makes the point from another direction: Australia pairs the highest open rate of any region with the highest unsubscribe rate, which MailerLite explains sensibly, since a reader has to see your mail before they can leave it.
Use an industry figure the way you would use a population average in medicine. It tells you whether your numbers are strange, not whether they are good. Two senders with the same rate can be in opposite kinds of trouble.
Does sending more often raise unsubscribes or lower them?#
The intuition is that more mail means more exits, and the published advice goes the other way. MailerLite, in the improvement section of its own benchmarks article, recommends sending campaigns more frequently to build a relationship with readers and sticking to a regular schedule. The logic is that a reader who sees you often enough recognises you, and recognition is what stops a message reading as unsolicited.
Both can be true, and the variable that decides it is whether the audience asked for this relationship. Going from four sends a year to four sends a month on a list gathered for a single event produces exits, because the frequency change is the only thing they remember about you. On a list that signed up expecting a monthly issue, irregular silence followed by a burst does more damage than the extra sends do.
The practical version of this is boring and reliable: pick the cadence you can keep, put it on the signup form, and then keep it. A rate measured against a promise nobody made is going to look mysterious whatever you do to the writing.
What should an unsubscribe page do?#
One click, and then stop. Every extra step between a reader deciding to leave and the list recording it converts an unsubscribe into a complaint, which is the outcome you cannot afford. A survey is the usual offender, and the compromise that keeps both sides is a single optional line with four choices, offered after the removal has already happened rather than as the price of it.
Two details decide whether the page is honest. First, the confirmation should say what changed in plain words, and must not re-add the address to anything. Second, if you offer a way to receive less instead of nothing, that option has to be visible on the same screen, because a reader who cannot find it takes the door rather than the window. Yahoo's published requirement that a one-click unsubscribe works and is honoured within two days is the floor for the mechanism, and a page that meets it quietly is worth more than a page that apologises at length.
A rate is only useful if the same denominator is used every send. Campaign Kicker reports each campaign with unsubscribes as a count and as a share of delivered mail, bounces and complaints counted beside it, and the address that leaves drops out of every later send on that account, including automations, without you editing a list.
Start free, no card