Most people ask this question expecting the answer to be a tool. Mailchimp or something else? In practice the tool is the last decision you make and the one that changes the outcome least.
A list that works is not a collection of email addresses. It is a group of people who agreed to hear from you and still remember why they agreed. The difference between those two things does not show on day one. It shows at month three.
Start with the right question: what are you offering?
For somebody to hand over an address, they have to get something back. That something needs to be specific enough to look useful and small enough that you can actually deliver it.
The common mistake is offering something too broad, like a weekly newsletter about marketing. Who signs up for that? Somebody interested in marketing generally, which means somebody who is not about to buy anything specific from you. The list grows and the result stays at zero.
The alternative is to offer one thing that solves one problem for one kind of person. A file that calculates something. A checklist for a task they repeat monthly. An explanation of the exact step they are stuck on. Whoever signs up for that has told you something about themselves you needed to know.
What to give away, and what not to
The rule I work to: give away the knowledge, not the execution. Explain the method completely, without holding pieces back. That builds trust and saves everyone time. What gets sold is the execution, the part where somebody actually does the work.
People who hold back the explanation in case nobody needs them afterwards lose twice. They lose the trust, and they lose the fact that the reader who works through the method and realises it is bigger than they thought is the best customer they were going to get.
The rules to know before the first send
This part gets skipped in almost every article about email marketing, and it is the part that, done wrong, ruins the whole list. Google publishes requirements for bulk senders. This is documentation, not my opinion.
Google's sender guidelines define a bulk sender as any sender who sends close to 5,000 messages or more to personal Gmail accounts within a 24 hour period. If that is you, conditions apply.
The documentation says senders should keep their spam rate below 0.1% and prevent it from reaching 0.3% or higher. That rate counts people actively pressing the report spam button, so one complaint in a thousand is the comfortable ceiling, not one in a hundred.
It also requires one click unsubscribe for marketing and promotional messages specifically, not for transactional ones like password resets, and it requires unsubscribe requests to be honoured within 48 hours. On the technical side it requires both SPF and DKIM, with the sending domain in the From header aligned to one of them, plus a DMARC record.
None of that is a detail for engineers. It decides whether your messages arrive or land in spam, which is the difference between a list with value and a list that is just a number.
Being small has its own rules
If you are starting out you are almost certainly nowhere near 5,000 messages a day, and that is an advantage. It means you can work with simple tools instead of paying a monthly subscription for a platform you do not yet need.
Google publishes hard quotas for Apps Script: 100 email recipients per day on a consumer account and 1,500 on Google Workspace. If your list is eighty people, that is comfortably enough and there is nothing to pay for yet.
Those numbers are useful for a second reason. They give you a clear line at which it is time to move to a real tool, so the decision gets made on a threshold rather than on a feeling or an advert you happened to see.
Where the form sits matters more than how it looks
A practical point that gets neglected: the signup form belongs where the reader is convinced, not where you find it most visible.
If you have written an article that explains something, the form belongs at the end of it, after the reader has finished and actually got something. A form that covers the screen three seconds after arrival meets somebody who has read nothing yet, and closing it is the reasonable response.
The wording beside the form matters more than its design. Subscribe to the newsletter says nothing. A sentence stating exactly what arrives, how often, and that they can leave whenever they like will raise signups and lower complaints later, and those are the two numbers you care about.
The mistakes that kill a list early
First, buying a list or scraping addresses from somewhere public. That is not a shortcut. It is the fastest route to the spam rate above, and it damages your sending domain in a way that is slow and awkward to repair.
Second, adding people without telling them what they will get and how often. Somebody who was not expecting to hear from you reports the message rather than unsubscribing from it.
Third, and by far the most common: building the list and then leaving it. Someone who signed up three months ago and heard nothing has forgotten you completely, and your first message arrives as an intrusion. If you are not going to write regularly, do not start collecting yet.
What the list works alongside
Email is not a channel that works on its own. It is where you move people who already found you somewhere else. Without a source bringing new people in, the list stays exactly the size it is.
That opens onto a bigger subject, which is how to grow without spending on ads, and I went through it in growing a business without paid advertising.
As the list grows the question shifts from collecting to organising. Automation starts to make sense at that size, and I set out the reasons in five reasons for email automation, and the shape of the funnel itself in the email marketing funnel.
If you run a shop, the first automation worth your time is the abandoned cart sequence, which I covered in abandoned cart emails.
Where this stops being true
One thing to say plainly: the rules above are Gmail's. Other providers such as Yahoo and Outlook have requirements that are close but not identical, and if most of your audience sits with one provider, read that provider's documentation rather than assuming.
Second, I have deliberately given no open rates or conversion benchmarks. Those numbers vary enormously by sector, list size and how the list was built, and most benchmark figures circulating online are averages across things that have nothing in common with each other.
Last, there are businesses where email is simply not the priority. If you sell something bought once in a lifetime, or your audience does not open email at all, the hours spent here are better spent elsewhere. If you want a structured route into the subject rather than assembling it from articles, colleagues at U Courses have written a beginner's guide that covers what changed in the rules: an email marketing course for beginners.