Email Preference Centre Template

Preference Options

Each option, the reason it exists, and exactly which sends it turns off. The stop column is the one that is usually empty, and an option that stops nothing is a checkbox that changes nothing.

Revenue per send rises as frequency falls, from $0.1140 on the full programme to $0.4200 at three sends a year. That is selection and not a treatment effect: the people who chose to hear less are the people who buy when they do hear. Read it the other way and you will conclude that mailing less earns more, which this table does not say.

Transactional sends stay on in every row, full unsubscribe included, and each option's copy says so. An option that suppresses an order confirmation produces a support ticket. The related check is the reverse one: a cross-sell inside an order confirmation is not transactional, and the primary-purpose test decides that by what the message contains rather than which system sent it.

The pause names a date. An indefinite pause is an unsubscribe with extra steps, and the retention data says a dated one is barely better.

Retention by Option

The first twelve months of unsubscribe clicks, 11,480 of them, each person followed 365 days. 9,184 reached the centre and 2,296 were held out to the old single-button page. Subscriber-days is the area under each survival curve.

The pause retains 94.1% at day 90, second best on the page, and 31.0% at day 365, worst on the page. Both numbers are true. The first one is true because the pause has not ended: it ends, the mail resumes, and 50.8% of the day-90 survivors are gone by day 180. Any window shorter than about 180 days ranks the weakest option as the strongest.

Retention and revenue rank the options in opposite orders. Three sends a year retains best, 85.9%, and earns $1.16 per person choosing it. The topic option retains 73.9% and earns $3.66, a 3.15x gap running the other way, because a retained subscriber receiving three emails a year is a retained subscriber receiving three emails a year.

The trade, stated: those 2,321 people would return $12,701 a year on the unchanged programme and return $9,186 on the settings they chose, 72.3% of it. The alternative was 2,321 people at zero.

The page itself

Preference Centre Copy, written for this reason profile. Options are ordered by the reason analysis, so the topic checkboxes come first and the frequency choice second.

Before you go, would less mail help, or different mail?

You can pick the topics you want, hear from us less often, or pause for three months. Or unsubscribe from everything, which takes one click below.

The full unsubscribe stays on the page because that is the condition on which the whole menu is lawful. A preference menu can be the opt-out mechanism CAN-SPAM requires, provided the menu includes an option to receive nothing at all. So demoting it to raise the save rate does not make the page unfriendly, it makes the menu something other than an opt-out mechanism.

Three common features fail 16 CFR 316.5 outright. A login wall requires information beyond the address and the preferences. A required reason field does the same. A confirm-then-confirm-again sequence is a second page, and the standard is visiting a single web page.

Deliberately absent: any discount, any guilt, any claim about what the reader will miss. An offer here converts an exit into a transaction and teaches the list that clicking unsubscribe produces a coupon.