How to get more email subscribers: where they come from

Find the stage that is losing people first

A sign-up number is the product of four stages, and improving the wrong one changes nothing. Count them downward: how many people see the ask, how many submit, how many confirm, and how many click the first message. The weakest link in that chain is the only place work pays, and it is usually not the place attention goes.

The commonest misdiagnosis is treating a traffic problem as a form problem. A page that converts well and is seen by forty people a month does not need a better headline; it needs to be linked from something. The second commonest is the reverse: pouring traffic onto a page that converts at 2 per cent because the promise is vague, which multiplies a bad number rather than fixing it.

So before changing anything, write the four figures down. If you cannot, that is the first job, and it is a day of work rather than a month. Everything below assumes you know which of the four is the low one.

Work the number backwards from a thousand

Start at the end and divide. A thousand confirmed readers is not a thousand sign-ups: under confirmed opt-in a fifth to a third of sign-ups never confirm, which is the range commonly cited in published double opt-in comparisons, so a thousand confirmed needs somewhere between 1,250 and 1,500 submissions.

Those submissions come out of visitors at the rate the placement converts: 15 to 30 per cent on a page whose only job is the offer, 1 to 3 per cent in a sidebar or footer, the planning bands explained under placement below. The same thousand readers therefore costs roughly 4,200 to 10,000 visitors through a dedicated page and 42,000 to 150,000 through a sidebar, which is the whole argument for building the page.

The nearest published figure is Unbounce’s median across 41,000 landing pages in all industries in the fourth quarter of 2024: 6.6 per cent, where the report itself says the definition of a conversion can vary greatly from one page to the next. Worked through the same thousand it asks for roughly 19,000 to 23,000 visitors, which is the last column of the table below. Read it as the middle of all landing pages, half of which convert below it, not as the figure for a page built for one offer.

Read those two numbers next to each other before deciding what to work on. Ten times the traffic is not a target; it is a reason to move the ask. Nobody reaches a thousand subscribers on a sidebar form unless they were already going to reach a hundred thousand visitors.

How long it honestly takes

At 500 visitors a month to the offer page, converting in the middle of its band, a thousand confirmed subscribers takes somewhere around a year. At 2,000 a month it takes a quarter. Both of those assume the page exists, converts in its band, and that confirmation is working — and the year is the more common case, because 500 visitors a month to one page is already more than most new sites send anywhere.

Growth also should not be faster than the sending reputation can carry. A domain with no history that suddenly sends thousands of messages looks like a compromised one to the large mailbox providers, which is why warming up over days and weeks is part of the plan rather than a caution. A list that doubles while its complaint rate doubles is not growing; it is being taken away a month in advance.

The useful reframing is that a thousand is not the milestone. A hundred people who asked for something specific and read it will reply, buy and refer at rates that a thousand indifferent addresses never reach, and they do not dilute the engagement signal that decides whether the next message arrives at all.

Make the offer something a reader can picture

The single largest determinant of a sign-up rate is what signing up gets somebody, stated concretely enough to evaluate in a few seconds. Join my newsletter asks the reader to guess and to trust the guess. Five emails that get your first paying client, one a day, about ten minutes each names a thing, a cost and an outcome, and a reader can decide.

This is not a copywriting trick, which is why it is worth more than the copywriting. The vague version asks for an open-ended commitment of attention with no stated return; the specific version asks for a bounded one with a named return. People decline the first and accept the second, and no amount of design work on the first turns it into the second.

What shape the offer takes is the other half of the decision, and the list-building guide sets out why a finite thing delivered by email does more for a list than a download.

Move the ask to where the intent already is

Placement changes the conversion rate by an order of magnitude, and it does so through intent rather than design. A page whose only job is the offer is planned at 15 to 30 per cent of the people who reach it, a band for one offer and visitors who came for it rather than a published measurement. The same form in a sidebar, a footer or the end of an article typically converts 1 to 3 per cent. A well-timed scroll or exit prompt sits between the two, usually in the mid single digits.

The reason is upstream of the page. Somebody who followed a link that named the offer had already made most of the decision. Somebody who reached the bottom of an article came for the article. Nothing on the form changes which of the two is looking at it, which is why the highest-leverage change is where the link to it sits.

In practice that means one dedicated page, linked from everything that mentions the subject: the end of relevant articles, the author line, the profile bio, the reply you send to questions about it. Sitewide placements are worth having and are not worth tuning; the dedicated page is worth tuning and is usually the one nobody has built.

Cut the form to one field

Ask for the address and nothing else. Every additional field is a deduction from the sign-up rate, paid in exchange for data nobody will use; what a first name field in particular costs, and the rare case where a second field earns its place, is in the list-building guide.

If you genuinely need a second piece of information, ask for it in the first message instead. A person who has confirmed an address and read something useful is far more willing to answer a question than a stranger evaluating a form, and their answer is more likely to be true.

The one thing worth adding to a form is a sentence rather than a field: what happens next, including that a confirmation message is coming and the name it will arrive from. That sentence costs nothing and recovers part of the confirmation loss, which is the stage most lists lose people at without noticing.

Recover the sign-ups lost at the confirmation click

Under confirmed opt-in, a fifth to a third of sign-ups never confirm, which is the range commonly cited in published double opt-in comparisons. That loss is partly structural and partly repairable, and the repairable part is usually the largest single number available to somebody who has already fixed the offer and the placement.

Four things recover it, and the confirmation email template writes out the message itself. Warn on the form that a confirmation is coming and from what name. Send it immediately rather than on a schedule. Put the link high in the message, as readable link text rather than a button image. And send exactly one reminder a day later, which commonly recovers a slice of the gap.

Then stop. An address that has received one message and one reminder has answered the question, and continuing is how a sender acquires complaints from people who never joined. The remaining loss is mistyped addresses, bots and thirty-second interest, none of which was ever going to read anything.

Widen the traffic last, one channel at a time

Only when the page converts in its band is more traffic worth buying or earning, and then the channels differ more in what they cost than in what they deliver. Writing that answers a question people search for compounds and starts slowly. Being a guest on somebody else’s list or show borrows trust and converts in the mid single digits of their audience. Advertising buys volume at a known price and buys no relationship at all.

Add one channel at a time and keep the complaint and bounce rates in view while you do. A channel that produces addresses which never engage is worse than no channel: it dilutes the engagement signal the mailbox providers read, and on a shared sending reputation the damage is not only yours.

The channel most often skipped is the cheapest. Individual messages to people who already know you convert better than anything purchasable, and they are lawful in a way that importing the same contacts is not. Ask; do not add.

In rough order of what each hour of work returns, for a list that already converts in its band:

• Linking the offer page from every piece of writing that mentions the subject. An afternoon, and it usually doubles the number of people who see the ask.
• Asking people you already know, one at a time. An evening, a small pool, the highest conversion you will ever see.
• Guest appearances on lists, podcasts and communities whose audience overlaps yours. Weeks of asking, mid single digits of each audience.
• Writing that answers a question somebody types into a search engine. Months before anything happens, then it compounds without a ceiling.
• Paid traffic to the offer page. Immediate, priced per address, and it stops the day you stop paying.

What is not worth doing, and what actively costs you

A long tail of tactics exists because they are easy to write about, not because they work. Button colour, arrival animations, countdown timers and longer sales copy on the offer page all move a sign-up rate by fractions at best, and a timer that resets on reload costs credibility outright.

Two things are worse than a waste. Buying or renting addresses produces complaints on the first send and reaches the spam traps seeded in every purchased file, which marks the sending domain directly. Giveaways that reward an entry rather than an interest produce addresses attached to nobody: the list gets bigger, the engagement rate collapses, and the mail stops arriving for the readers who did want it.

The honest summary is that there are three big levers and a very long tail. Pull the three, measure, and spend the remaining time writing something worth subscribing to.