Newsletter from Zero: Growth, Content, Automation

Social media doesn't belong to you. A post's reach is set by an algorithm that can change without warning, and an account you spent three years building can lose half its reach in a week. With a newsletter, you own the contact: your list of email addresses is your asset, and nobody stands between you and a reader deciding whether a message gets shown. It's the slowest-growing, most reliable channel you can build.

The downside is just as honest: a newsletter is work you can't fake your way through. You need a reason for someone to hand over their address, something to send them regularly, and enough technical hygiene that it doesn't end up in spam. AI won't take that work off your hands, but it strips out the mechanical half of it — lead magnet ideas, issue skeletons built from your notes, subject line variants, reading the numbers from an export.

This guide moves from the first decision (“what's it about, and for whom”) to steady, ongoing operation. Every phase comes with a prompt to copy. And two rules apply throughout: only a human hits send — a mass email can't be unsent, and AI never gets to release one on its own — and your voice is the one thing that sets your newsletter apart from every other one, so the model learns to write like you, not the other way around.

A typical scenario

Marek is a process consultant for manufacturing. He spent three years posting on social media, built up over four thousand followers, and got zero clients out of it. When his post reach dropped to a tenth of what it was, it hit him that he had nothing — not even a list of people he could actually email.

He started from a single question: what could he give people every week without it turning into a second job. The answer was “one specific shop-floor problem and how it gets solved” — ten minutes of reading. Instead of the usual “subscribe for updates,” his sign-up offer was a checklist he already used with every client. He pulled it out of his own notes in an afternoon.

He wrote the welcome sequence once: four emails that go out automatically to a new subscriber over the first two weeks. He writes each issue by dictating raw notes from the week, having a draft built from them using his style profile, and then spending an hour rewriting it. Send day is Thursday morning, always.

A year in, he has nine hundred subscribers — fewer than he had followers. But out of those nine hundred came eleven leads, because an email from someone people have read for fifty straight weeks carries a weight a post never will. And when the platform changes its algorithm again next year, it won't touch him.

Phase 1: deciding what the newsletter actually is

The most common reason newsletters die after the third issue isn't a lack of time. It's a decision that never got made about what the newsletter is supposed to be — and without it, writing every issue means deciding from scratch again.

Promise, rhythm, scope

Three things to decide up front and then stick to. The promise is one sentence about what the reader gets and what it does for them: “every Thursday, one specific manufacturing problem and how to fix it.” The rhythm is a frequency you can hold even in your worst week — every two weeks, kept, beats weekly, skipped, by a mile. And the scope is a length you can write and a reader can finish; ambitious twenty-minute essays end up unread.

What a newsletter isn't: a list of your own updates (“what's new with us”), a channel for pushing an offer into every single issue, or a digest that forwards other people's articles with no opinion of its own. All three get unsubscribed from at about the same rate.

I'm about to start a newsletter and I need this to be a decision,
not a vibe.

What I do: [field, who I help, how I make a living]
Who I want to write to: [audience description and their situation]
What I can write about long-term: [3–5 topics]
How much time I have per issue: [e.g. 2 hours]
What I want to happen next: [leads / course sales / a community]

Propose 4 different newsletter concepts. For each:
- The promise in one sentence (what the reader gets and what it
  does for them)
- Exactly who it's for and who it's not for
- Rhythm, and an estimate of how much work one issue takes
- Structure of a single issue (which sections, and why)
- How it differs from what already lands in my industry's inboxes
- Where it breaks down after twenty issues (what will wear thin)

Take the last point seriously — I want to see what's sustainable.
Don't invent facts about my field; work only with what I wrote.

You'll get back four options where the difference in workload is visible. Choose based on the last point, not on what sounds best — sustainability decides everything else. Save the concept you pick into a project with persistent context; you'll come back to it for every issue.

Phase 2: lead magnet and sign-up form

“Subscribe for updates” is an offer nobody responds to. People hand over an address for something specific — and the narrower it is, the better it works. Nobody wants a generic ebook about productivity; a checklist for handing off a project to a client, on the other hand, does.

Three formats that work

A checklist is the cheapest to produce and has the best value-to-effort ratio: one page someone prints and checks off. A template — a document, spreadsheet, or ready-to-copy text — works even better because it saves work immediately. A mini-course split across a few emails has the added benefit that a reader gets to know you gradually, not through a single download.

Across all three, one rule holds: a lead magnet has to solve one small problem completely, not a big problem halfway. And it has to connect to what you actually do — otherwise you attract people you have nothing else to offer.

I want to build a lead magnet — something I give people in exchange
for their email.

My field and what I sell: [description]
Who I write to: [audience]
Newsletter promise: [promise from phase 1]
What I already have: [source material, notes, slides, processes,
client replies, spreadsheets I use]

Propose 5 lead magnets. For each:
- A specific title (not “the guide to X”)
- What ONE problem it solves, and for whom
- Format (checklist / template / mini-course / calculator)
- What existing material it can be built from
- An estimate of how much work it takes
- Why someone would give up their email for it — and how it beats
  what people can find for free with a search

Rank them by value to the reader relative to my effort.
Don't propose anything I'd have to build from scratch over a
whole week.

You'll get back suggestions built on what you already have. Check the “free competition” point yourself — the model has no idea what's standard in your field and can easily suggest something that's already free on the first page of search results.

Producing the lead magnet's content

Once you know what it is, the rest is craft. This prompt handles more than half the work; the rest is layout.

I'm building a [checklist / template] called [name] for [audience].
It's meant to solve this problem: [description].

My source material — how I actually do this:
[paste notes, transcripts, old client emails]

Build the content:
- Step-by-step structure, in the order it's actually done
- For each step: what to do (imperative, one sentence), how to
  know it's done, and the most common mistake
- Cut steps that are obvious, and don't pad it out
- End with 3 check-yourself questions

Work only from my source material. Where a step is missing from
it, write FILL IN and exactly what I need to supply.
Write in my language, not the language of marketing guides.

You'll get back a skeleton to finish and lay out. The FILL IN lines matter more than the finished parts: they show where your process is unspoken, and that's exactly where a lead magnet tends to fall apart.

Form and double opt-in

The technical part is short, but it decides the quality of your list. Put the form where someone is already reading something of yours — the end of an article, a footer, a dedicated page you link to. The form's label should say what arrives and how often, not “subscribe to our newsletter.”

Double opt-in means that after filling out the form, a confirmation email goes out and someone only becomes a subscriber once they click it. You lose some share of sign-ups and gain three things in return: typos and someone else's address planted as a prank are filtered out, your list only holds people who actually want to be on it, and you have provable consent. Turn it on — it's a single checkbox in your sending tool's settings.

Phase 3: welcome sequence

Interest peaks the moment someone signs up. If nothing arrives for a month, they've forgotten who you are by the time the first regular issue lands. A welcome sequence is a series of emails you write once, and the tool then sends automatically to every new subscriber.

A proven skeleton is four emails over ten days: the first delivers the promised lead magnet and says what's coming next (sent immediately). The second, two days later, shares your best existing content on the topic. The third, four days later, explains who you are and why you know this stuff — a story, not a resume. The fourth, a week later, asks one question and invites a reply; replies are gold, because you can write your next issue from them and they help deliverability too.

Write a 4-email welcome sequence for new newsletter subscribers.

Newsletter: [promise and rhythm]
Audience: [who I write to]
Lead magnet: [what they received]
Who I am: [3–4 sentences on what I do and why I'm good at it]
Best content I can point to: [links and what they're about]
Sample of my writing: [paste 300–500 words of my own text]

For each email, give:
- when it sends (day after signup)
- subject line (max 45 characters) and preheader (max 90
  characters)
- body copy written in my style, per the attached sample
- one call to action; across all four combined, no more than one
  mention of what I sell

Rules:
- No “hope this finds you well” and no addressing a crowd.
- Email 4 should end with one specific question to the reader.
- Don't invent anything about me — use only what I wrote.

You'll get back a sequence that, once rewritten into your own voice, goes into your tool and is done for a year. Check that email one actually contains the link to the lead magnet — the most embarrassing mistake in email marketing is a welcome email missing the file it promised.

Phase 4: writing issues in your voice

This is where the entire value of a newsletter lives. Text that sounds like it came out of a generator is obvious by the second paragraph, and nobody finishes it. The way to get AI writing in your voice doesn't run through asking it to “write naturally” — it runs through samples of your own writing.

A style profile: have it describe how you write

Build a description of your own style once, then paste it into every prompt. It works better than any instruction you could write ahead of time.

Here are 5 of my own texts that I consider good and typical of me:

[paste texts, 2,000–4,000 words total]

Describe my writing style precisely enough that someone could use
it to write a new piece that's hard to tell apart from mine.
Specifically:
- sentence and paragraph length, rhythm, how I mix short and long
- how I address the reader (formal/casual, individual/crowd)
- words and phrases I use repeatedly
- words and phrases that are NOT in my writing
- how I open and how I close
- how I use examples, numbers, and admitting uncertainty
- how much humor, and exactly what kind

Write the output as instructions for another writer, not as
analysis. Attach an example from my texts for every point.
Don't flatter me — describe what I do poorly or repetitively too.

You'll get back a profile worth saving — you'll use it for every issue and for sales copy too. Pay special attention to the “not in my writing” list: it's usually the most valuable part, because it's exactly the phrasing AI tends to stuff into drafts (see showing AI a sample).

An issue draft from raw notes

Never prompt with “write a newsletter about [topic]” — you'll get text anyone could have written. Feed it your own material instead: notes from the week, a dictated voice memo, fragments from client calls.

I'm writing an issue of my newsletter. Here are my raw notes on
the topic — unorganized, exactly as they occurred to me:

[paste notes or a dictation transcript]

My style profile: [paste the profile from the previous step]
Issue structure I stick to: [sections]
Length: [e.g. 700 words]

Build a draft:
- Keep my ideas and my examples, don't add new ones
- Put them in a logical order and add transitions
- Open with a specific situation, not a general statement
- Where a note is unclear, write QUESTION: and ask
- Where a claim would need support I don't have, write VERIFY
- Don't generalize into platitudes and don't add a summarizing
  conclusion that just repeats what was already said

At the end, list every QUESTION and VERIFY.

You'll get back a draft with your thoughts put in order — and, more importantly, a list of the spots you never fully thought through yourself. Rewriting the draft in your own words is still a required step; treat it as an outline, not as text ready to send.

Sections and rhythm

A fixed structure saves more time than any prompt. When you know an issue always has three parts, you're not deciding from scratch every time. A proven skeleton: one main thing (the issue's topic, two-thirds of the text), one small thing to use right away (a tip, a shortcut, a phrase), and one link with your own comment (why it's worth reading). Add a fixed send day and time — readers get used to the rhythm faster than you'd expect.

A backlog of topics is what decides whether you keep going. Keep a running list of ideas and add to it the moment something occurs to you, not when it's time to write.

Subject line and preheader

The subject line decides whether it gets opened, and it's the one place worth writing ten variants for. The preheader (the text that shows after the subject in an inbox list) isn't a second subject line — it complements the first one.

Here's a finished issue of my newsletter: [paste text]

Write 10 subject line variants, each max 45 characters, and for
each a preheader up to 90 characters that complements the subject
without repeating it.

Vary the approach across them:
- a specific detail from the text (a number, a tool name, a
  situation)
- a question the text answers
- a contradiction or a surprising claim from the text
- a plain, descriptive subject with no hook

Rules:
- No all caps, exclamation points, emoji, or words like “free,”
  “last chance,” “you won't believe” — they trip spam filters and
  kill trust.
- Nothing the text doesn't deliver on. A subject line is a
  promise, not bait.
- One sentence per variant on who it's aimed at.

You'll get back material to pick one from and maybe tweak. Testing subject lines only starts to mean something on larger lists — on two hundred addresses, the difference between variants is indistinguishable from chance.

Check before sending

A mass email can't be taken back. This check costs five minutes.

Review this text before I send it as a newsletter issue. Don't
rewrite it, just list the findings with the location quoted:

[paste the issue]

1. Factual claims, numbers, and names that need verifying against
   a source (don't judge whether they're true — just list them).
2. Passages that read as generic AI writing: stock phrases,
   unsupported superlatives, empty summarizing sentences.
3. Promises the text doesn't deliver on, and mismatches with the
   subject line.
4. Links: missing, broken-looking, undescribed, unclear where
   they go.
5. Personal or client information that shouldn't go out.
6. Sentences a reader without context won't understand.

Always resolve point 1 against the source, not by asking the model — the process is in the tip on fact-checking with AI. And whatever else you check, the last step is non-negotiable: send yourself a test email and read it on your phone. A human sends, always — never a routine.

Phase 5: segmentation and measurement

Segmentation by behavior, not by gut feel

Segmentation means splitting your list into groups you send different things to. It only makes sense when it actually changes something — otherwise it's needless complexity that slows down your writing.

Three segments worth setting up even on a small list: by sign-up source (who came through which lead magnet, because that tells you what they care about), by clicked topic (who clicked a link on a given topic wants more of it), and inactive (who hasn't opened or clicked in six months). The last group matters most and is handled in the deliverability phase.

Here's an export of my subscriber list — columns and a few rows
(no email addresses, just behavior and attributes):

[paste header and sample rows]

I send [rhythm] and I want [goal: more leads / better
deliverability / course sales].

Propose 3 to 5 segments that make sense for a list my size. For
each:
- the condition that defines it (specific, based on the columns
  above)
- what I'd send that segment differently
- roughly how many people would be in it, based on the data provided
- what goes wrong if I add this segment (more work, a small sample)

Don't propose a segmentation where I'd end up sending everyone the
same thing anyway.

You'll get back a proposal built on what you actually collect. The last line of the prompt is deliberate: the most common beginner mistake is fifteen segments and one piece of copy for all of them.

What to measure now that open rate has lost its meaning

Open rate is unreliable today. Privacy protections in mail apps pre-load images on the user's behalf, so an open gets counted even for someone who never actually saw the email. The number is inflated, inconsistently across different groups, and comparing two issues by it is fortune-telling.

Track numbers backed by real action instead:

• Clicks — how many people clicked a link. The best available stand-in for “read it and it landed.”
• Replies — the strongest signal an email has. A few replies per issue mean you're writing for people, not into a void, and it helps deliverability too.
• Unsubscribes — normal; watch for spikes. A sudden jump means you changed the tone, the frequency, or the promise.
• Spam complaints — the most serious metric. Even a small share can hurt delivery of every issue that follows.
• List growth — new subscribers minus unsubscribes; the one number that should trend up over the long run.
• Conversions — what happened next: a lead, an order, a sign-up.

When you want to compare numbers across issues, the same rule applies as in data analysis: the model doesn't calculate, the model writes a script.

I have a CSV export of my issue statistics. Header and first 3
rows (semicolon delimiter, UTF-8 encoding, decimal comma):

[paste rows]

Write a Python script (pandas) that:
1. loads the file and handles the number formatting,
2. calculates click rate, unsubscribe rate, and complaint rate per
   issue, relative to delivered, not sent,
3. compares issues against each other and flags any more than one
   standard deviation off the average,
4. calculates net list growth by month,
5. prints the issue with the most unsubscribes and the one with the
   most clicks,
6. saves the result to results.csv.

For each step, write a comment explaining WHY it's done that way.
Don't calculate anything in your reply, I want a script to run
myself.

You'll get back a script that gives the same numbers every time. Make sure the rates are calculated from delivered emails — from sent, the ratios look better and mean nothing. And on small lists, take the deviations with a grain of salt: five unsubscribes out of two hundred is a difference a single issue's topic can produce on its own.

Phase 6: deliverability, consent, and upkeep

The best newsletter is worthless if it lands in spam. Deliverability rests on a handful of boring things you set up once and then just maintain.

The basics that have to be true

Your own domain and sender verification. In your sending tool's settings, go through the domain verification wizard — this sets up SPF, DKIM, and DMARC records that tell mail providers you're allowed to send on behalf of your domain. Without them, major providers push you to spam regardless of content.

Double opt-in, and never a purchased list. A list bought from someone else can wreck your sender reputation in a single send. Also don't add people just because they emailed you once — consent doesn't get inferred.

One-click, visible unsubscribe. A hidden unsubscribe link doesn't stop someone from leaving, it just turns the departure into a spam complaint, which is much worse for you.

List cleaning. Twice a year, remove addresses that haven't opened or clicked in a year, and continuously remove ones that keep bouncing. It hurts, because the list shrinks — but sending to dead addresses damages deliverability for everyone else on the list. Before removing someone, try one “still want to hear from us?” email.

Build me an operating checklist for my newsletter.

Tool: [Brevo / Mailchimp / other]. Domain: [my own / third-party].
List size: [count]. Rhythm: [frequency].

Build three checklists:
1. ONE-TIME SETUP — what to configure now (domain verification,
   double opt-in, unsubscribe link, sender address, footer with
   identification)
2. BEFORE EVERY SEND — reviewing the issue, checking links, test
   email, mobile check, correct segment
3. QUARTERLY — cleaning out inactive subscribers, checking bounced
   addresses, reviewing the welcome sequence, checking the stats

For each item, note where it's configured in the tool and how I'll
know it's working. Make each checklist fit on one screen.

You'll get back an operating document you can pin next to your monitor. Add one thing the model doesn't know: the last time you cleaned the list.

Consent and GDPR

Email addresses are personal data, and a newsletter is processing them. The rules aren't complicated, and following them mostly means staying well clear of the line:

• Consent has to be freely given, specific, and provable. A checkbox must never come pre-checked, and signing up for a newsletter must never be a condition of something else. Double opt-in also happens to be the best available proof that consent actually occurred.
• The person must know what they're signing up for: who the data controller is, what you'll send, how often, and how to unsubscribe. A link to your privacy policy belongs next to the form.
• Keep records. Store when and where someone signed up; sending tools usually log this automatically.
• Your sending service is a data processor. Providers usually have a signed data processing agreement ready in your account settings.
• Unsubscribe means it's over. Without undue delay and with no conditions attached.

I'm setting up a newsletter sign-up form and want the consent and
disclosures done properly.

Who I am: [sole proprietor/company, name, location]
What I'll send: [content and frequency]
What else I do with the address: [nothing else / behavior-based
segmentation]
Tool: [sending service and where its servers are, if I know]
Lead magnet: [what they get after signing up]

Propose:
1. Copy for the form (what arrives, how often, how to unsubscribe)
2. Wording for the consent checkbox
3. Paragraphs about data processing for the privacy policy page
4. Confirmation email text for double opt-in
5. A list of questions I should take to a lawyer, because I can't
   decide them from a generic template

Write in plain language, not legalese. Don't present this as legal
advice.

You'll get back a usable starting point and, more importantly, a list of questions. Take the last point seriously: the model isn't a lawyer, and whatever wording it generates is a draft to be reviewed, not a document ready to deploy. That matters even more the moment you're collecting more than just an email address.

Case study: this site's newsletter

The Produktivní.cz newsletter runs on Brevo and is deliberately simple — the whole mechanism fits in a few lines.

The sign-up form adds a contact to the “newsletter” list and, alongside the address, stores an attribute for the sign-up source: it shows whether someone signed up in the footer, under an article, or somewhere else on the site. It's the cheapest segmentation there is — one field that tells you something about a reader without asking them a single question.

The lead magnet is the ebook “Top 30 tips that give you back an hour a day” — exactly what phase 2 recommends: a ready-made selection from content that already existed on the site, not a new piece written from scratch. The welcome email goes out immediately after signup, contains the download link, and one sentence about what's next: one tip every Thursday. Requests for training go into a separate list with its own tag, so they don't get mixed in with newsletter readers — different intent, different communication.

One technical detail worth noting shows up in the form's code: if the welcome email fails to send, the signup itself doesn't fail. The contact is saved and the person gets a confirmation; the email can be caught up on later. The reverse order would mean a temporary outage costing you a subscriber who already said yes. The form also never fakes success — when the service is unavailable, it says so.

And what that means for you: start with one list, one lead magnet, one welcome email, and a fixed send day. Add segments, automation, and subject-line testing only once you have enough subscribers to actually test against.

The most common mistakes

• Collecting addresses with no reason. “Subscribe for updates” convinces nobody. Without something specific in exchange for an email, a list grows at a handful of people a month.
• Letting AI write a whole issue from a topic name. You get text without you in it — generic, interchangeable, obvious by the first paragraph. The input has to be your notes, not a topic assignment.
• Skipping the welcome sequence. Interest peaks on the day someone signs up. If nothing arrives for a month, the first real issue reads like it's from a stranger.
• Sending irregularly. A newsletter that shows up once a quarter doesn't get remembered, and people unsubscribe. A smaller frequency kept is better than a big one promised.
• Chasing open rate. The number is inflated by privacy protections in mail apps. Decide based on clicks, replies, and unsubscribes.
• Ignoring cleanup and deliverability. Unconfirmed addresses, purchased contacts, and a year of inactive readers hurt delivery for everyone else.
• Leaving sending on autopilot. A routine may prepare the draft. A human who read the whole thing presses send.

The best tools

• Brevo — sending, lists, welcome-sequence automation, and forms; this site's newsletter runs on it.
• Claude (claude.ai) — the style profile, issue drafts from your own notes, subject line variants, pre-send checks; Projects hold the newsletter's context.
• Claude Cowork — working across a folder of notes, transcripts, and drafts in progress, with nothing to copy-paste into chat.
• A voice recorder or dictation mode — raw notes for an issue get collected fastest by talking; the transcript then becomes the draft's input.
• Scheduled tasks — a “pull together material for this issue” reminder on a fixed day, so writing doesn't depend on motivation.

What you get out of it

• Time: an issue goes from “a whole evening” to an hour and a half — the style profile and drafts from notes cut out the mechanical half of the work.
• Peace of mind: a welcome sequence written once keeps working on its own, and a new subscriber never falls into silence.
• Quality: text written in your voice from your own notes can't be mistaken for anyone else's — and that's the only real defense against a newsletter blending into everyone else's inbox.
• An asset: your contact list is the one part of your audience nobody can take away from you or cut off with an algorithm change.

Pro tip

The best source of topics isn't ideas, it's your readers' replies. Ask one question per issue and save the replies to a single folder, and after six months you have a backlog of topics you know matter — because someone asked. Have it mapped out for you: “go through these replies and list the recurring questions, with an exact quote and a count of how many times each came up.” That same folder, incidentally, is also where you'll draw copy from if you ever run an ad campaign.

And the closing rule: rhythm beats perfection. Fifty average issues sent on time do more for your standing than five brilliant ones with six-month gaps between them.