How do I build an ecommerce email list?

Almost every address you will collect this year is going to come from someone who was already on your site. That reframes the job: list building is a conversion problem on existing traffic, not a new channel to fund.

So do the arithmetic before you start. Take your monthly sessions and the share of visitors who currently sign up. Lifting that share, even modestly, adds addresses every month with no extra media cost, and those addresses then produce visits for free. This is why we treat email capture as part of the visitors lever rather than as an afterthought to a send schedule.

It also sets the priority. If your site gets very little traffic, list building is not the constraint and you should be working on traffic. If your site gets real traffic and captures almost none of it, that gap is one of the cheapest things in the business to fix.

And it explains why buying lists never works. A purchased address has no relationship with you, never asked to hear from you, and damages the delivery of every message you send to the people who did ask.

The offer has to be worth more to the visitor than the mild cost of handing over an address, and a discount is only one way to get there.

A discount works, and it has a cost. You are giving margin away to people who may have bought anyway, and you are training the list to expect a code. If you use one, use it on first orders only, keep it small, and be aware you have set an expectation you will have to live with.

The alternatives are often stronger and cheaper. Early access to a product that sells out. A back-in-stock alert, which is the highest-intent capture on any store and takes a single small form. A size, fit or compatibility guide for a category where getting it wrong means a return. A short quiz that recommends the right product, which collects the address and improves the first purchase at the same time. For consumables, a reminder when it is time to reorder.

Match the offer to where the person is. Someone on a product page for an out-of-stock size wants the alert. Someone reading a guide wants the guide. One generic popup on every page is why most stores' capture rate is low.

Use several capture points rather than one popup, because different visitors are ready at different moments.

The ones that carry most of the volume: a site-wide sign-up shown after the visitor has done something rather than on arrival, which respects the person and usually performs better. The checkout, where a properly labelled marketing consent option collects buyers who are already giving you their address for the order. Back-in-stock and notify-me on out-of-stock variants. The footer, which collects the small number of people who go looking. And the post-purchase page and email, where you can ask for SMS consent separately from email.

Two details make a large difference. Ask for as little as possible, which for most stores is the address alone, adding a first name only if you will actually use it. And do not show the same prompt to someone who has already signed up or already bought, because that is the fastest way to make a returning customer feel like a stranger.

On mobile, check the form actually works with the on-screen keyboard up. A capture form that hides its own submit button behind a keyboard is a common and invisible loss.

Collect a clear, recorded opt-in for marketing, kept separate from the transactional messages about someone's order. This protects you legally and it protects the thing that actually matters day to day, which is whether your emails arrive.

The practical rules are consistent across most markets. Marketing consent is a positive action, not a pre-ticked box hidden in a checkout. It has to be clear what the person is agreeing to receive. Every marketing message needs a working unsubscribe that takes effect quickly. And you need to be able to show when and how consent was given, which good email platforms record automatically.

SMS is stricter than email nearly everywhere, and consent for SMS is separate from consent for email. Collecting one does not give you the other.

Then there is the technical side that decides whether your mail reaches an inbox at all. Authenticate your sending domain, send from your own domain rather than a free address, and warm a new sending domain up gradually instead of mailing your whole list on day one. None of this is marketing work, and all of it decides whether the marketing work is seen.

Track how many people open and click, and how much revenue the list produces, rather than the total number of addresses. A list that grows while engagement falls is getting worse, not better.

The numbers worth watching monthly: sign-up rate as a share of sessions, which tells you whether the capture points are working. Active subscribers, meaning people who opened or clicked in the last few months, which is the real size of your list. Revenue per subscriber. And the unsubscribe and complaint rates on each send, which are your early warning that something is wrong with frequency or relevance.

Clean as you go. Suppress addresses that have not engaged in a long time after one attempt to re-engage them, and remove hard bounces immediately. This feels like throwing away an asset and it is the opposite: inbox providers judge you on how the people you mail respond, so carrying dead addresses lowers delivery for everyone else on the list.

And remember what the list is for. It is not a broadcast channel for discounts. It is the only audience you own outright, and the cheapest source of returning visitors any store has.