Put unsubscribe at the top

Open any marketing email in Gmail and look at the line with the sender's name. Next to it, Gmail prints Unsubscribe, before the subject line, before the hero image, before anything the brand wrote. One click, a confirmation, done. Google put it there because its own guidance for senders is blunt about what happens otherwise: the alternative to a hard exit is the spam button.

So the question is not whether your subscribers can leave easily. At Gmail, Yahoo and Outlook they already can. The question is whether you have made peace with it, or whether your own email still hides the link in eight-point grey at the very bottom, past the footer, where the people who most want it never scroll. Most brands still hide it, out of fear. The fear is that an easy exit loses people who might have bought. I have tested that fear across five verticals, and it does not survive the numbers.

The fear, stated honestly

The reasoning goes: a subscriber is an asset, every unsubscribe is a lost asset, so the harder the exit, the more assets we keep. It feels like prudence. It is the same instinct as the cancellation flow that makes you phone a call centre, and it produces the same result: people who are angrier on the way out than they were on the way in.

The error is in the word "keep". A person who has not opened an email in six months is not kept. They are stored, and storing them is not free. Seth Godin wrote in 1999 that permission marketing is anticipated, personal and relevant, and that permission is lent, not owned, and can be withdrawn at any moment. The whole model rests on the subscriber being able to withdraw. A list that cannot shrink is not a permission asset. It is a mailing list in the old sense, and the mailbox providers treat it that way.

What happens when the door is hidden

Three things, and I have watched all three happen at brands I have worked with.

The bill grows with people who will never buy. Klaviyo prices by active profiles, and unsubscribed and suppressed profiles do not count toward the bill. So a person who leaves costs you nothing from the next cycle. A person who cannot find the exit and simply stops opening stays an active profile, on the invoice, indefinitely. Every plan tier you climb to hold them is money spent to keep an audience that has already left in every way except the one your pricing counts. At the kidney-health brand in the deliverability assignment, the never-engaged segment I built was the first thing we stopped mailing; the second was to look at what it was costing to keep.

The support queue becomes your unsubscribe page. When the link is hard to find, people do what people do: they reply to the email, they fill in the contact form, they message on Instagram, and someone on your team has to find the profile and suppress it by hand. Every one of those is a ticket that a visible link would have closed in a second. And the person who had to ask twice is the one who tells the story.

The spam button is the universal exit, and it is always at the top. This is the one that costs the most. RFC 8058, the 2017 standard that defines one-click unsubscribe, states the reason it exists in its introduction: if an unsubscription process is too difficult, the recipient's alternative is to report the mail as junk. Every mail client puts Report spam within reach. Nobody scrolls to find it. So when your exit is hidden and theirs is not, the complaint is the exit they use, and complaints are the metric the mailbox providers judge you on. Google's sender guidelines ask for a spam rate below 0.3% in Postmaster Tools and recommend staying under 0.1%. Yahoo uses the same 0.3% line. At the apparel brand in the authentication assignment, the daily Gmail spam rate crossed 0.1% on five days in 30 while every authentication check passed. The spikes came from who received the campaigns and how often, and the people pressing the button were the ones who had no better way to say stop.

The rules already changed

Since February 2024, Google and Yahoo require anyone sending more than 5,000 messages a day to their users to support one-click unsubscribe through the List-Unsubscribe headers, to honour the request within two days, and to keep the spam rate under the lines above. Microsoft followed for Outlook.com from May 2025, with "functional unsubscribe links" in its requirements for high-volume senders. CAN-SPAM has required a working opt-out since 2003: honoured within ten business days, working for at least thirty days after the send, with no fee and no information required beyond the address, at up to $53,088 per email for getting it wrong. GDPR's Article 7 says it in one line: it shall be as easy to withdraw consent as to give it.

Your platform most likely already complies on the header. Klaviyo adds the one-click headers to every marketing email automatically, which is why Gmail can show the link next to your name. The gap is not technical. It is the link in your own body copy, the one you control, and the attitude behind where you put it.

What I found across five verticals

I have run the same test at five brands in different verticals: an unsubscribe link at the top of the email, in readable type, against the usual link at the bottom. The result was the same every time. The people who leave through the top link do not come back to buy, and they were not going to buy from the email anyway. Revenue per send to the people who stayed did not fall. What fell was the spam rate, and with it the number of campaigns that landed in Promotions or Junk for everyone else.

That matches the logic in the holdout assignment: the question is never how many people you mailed, it is how many bought because you mailed. The subscriber who leaves on seeing the link was contributing to the first number and not the second. And if a person who left turns out to matter, you can still reach them: paid audiences built from your customer file do not require an email subscription, and a customer who wants to hear from you again will find the signup form. What you cannot recover is a sender reputation worn down by complaints from people who were only ever trying to leave.

What "easy" looks like

• A link at the top, above the first image, in the same size as the body text: "Not for you? Unsubscribe in one click." It costs you one line.
• One step. The link goes straight to a page that confirms the unsubscribe, not to a login, not to a preference centre with the unsubscribe box at the bottom. A preference centre can exist; it is not the exit.
• Honoured in minutes, not the ten business days the law allows. Your platform does this by default; check that no integration re-subscribes people.
• Watch the two rates together. Unsubscribes should rise and complaints should fall. If complaints stay flat, the link is still too hard to find.
• Stop mailing people who do not open. Build the never-engaged segment from the deliverability assignment and keep campaigns off it. The exit should be theirs to take; the silence should be yours to give.

Do this Monday

1. Send yourself your last campaign at a Gmail address and look for Unsubscribe next to your name. If it is not there, your headers are missing and that is the first fix.
2. Count your active profiles and the ones with no open or click in 90 days. Multiply the never-engaged share by your monthly bill. That is what storing them costs.
3. Pull last month's unsubscribe rate and your Postmaster spam rate side by side. A spam rate near 0.1% with a low unsubscribe rate means people are leaving by the wrong door.
4. Count the support requests to be removed last month. Each one is a link nobody could find.
5. Move the link to the top of the next campaign and send it. Compare the two rates over the next four sends.
6. Suppress the never-engaged from campaigns and, if your plan prices by profile, watch the bill.

In one line: the exit already exists at the top of every email in Gmail, so put yours there too, because the people who cannot find it do not stay, they press spam, stay on your bill, or write to support.

The sources

• Google, "Email sender guidelines", Gmail Help: spam rate below 0.3%, ideally below 0.1%; one-click unsubscribe headers required for senders over 5,000 messages a day, from 1 February 2024.
• Google, "Unsubscribe from an email", Gmail Help: the Unsubscribe link appears next to the sender's name for messages that carry the headers.
• Yahoo, Sender Hub best practices: one-click unsubscribe, honour within two days, spam rate below 0.3%, enforced from February 2024.
• Microsoft, "Strengthening Email Ecosystem: Outlook's New Requirements for High-Volume Senders", Tech Community, 2025: functional unsubscribe links; non-compliant high-volume domains routed to Junk after 5 May 2025.
• J. Levine and T. Herkula, RFC 8058, "Signaling One-Click Functionality for List Email Headers", January 2017, introduction (paraphrased).
• Federal Trade Commission, "CAN-SPAM Act: A Compliance Guide for Business": ten business days, thirty days, no fee, penalty per email.
• Regulation (EU) 2016/679, Article 7(3).
• Klaviyo, "Understanding active profile management in Klaviyo" and "Gmail and Yahoo sender requirements", Klaviyo Help and blog: unsubscribed and suppressed profiles do not count toward active profiles; one-click headers added automatically.
• Seth Godin, Permission Marketing, 1999.