What we run
Newsletter growth, run for you.
Newsletters are two channels wearing one name. Running your own is a writing and shipping problem that compounds slowly. Getting placed in someone else's is a relationship problem, closer to creator work than to content. Zway drafts and ships your issues, answers the replies, and negotiates the placements and swaps by hand.
Last reviewed 27 August 2026
The failure mode
What breaks when AI does this alone.
Your own newsletter fails quietly. An automated send goes out on schedule, nobody replies, and the list decays while the dashboard still reports a successful delivery. The replies are the entire point of the channel and they need a person to answer them. Placement in other people's newsletters fails more loudly. A cold automated pitch to a writer who has never heard of you gets ignored, a second one gets you filtered, and writers overwhelmingly place people they already know.
What our operators do here
- Draft each issue in your voice and cut the parts that only sound good to you
- Ship on the same day every week, including the weeks when nobody feels like it
- Answer every reply personally, because the replies are where the pipeline comes from
- Pitch guest slots, swaps, and sponsorships to writers by hand, one at a time
- Negotiate placement terms and confirm what the copy will actually say
- Keep deliverability clean, prune dead addresses, and watch the sender reputation
- Report which issues produced replies and which were filler
Volume
What ships in a typical week.
| What | Typical volume |
|---|---|
| Issues drafted, edited, and sent | 1 |
| Replies answered by a person | All of them |
| Placement or swap pitches sent by hand | 5 to 10 |
| Placements negotiated and scheduled | 1 to 3 per month |
| List hygiene and deliverability checks | Weekly |
Volumes are typical rather than contractual. They move with what the channel is actually returning.
What good looks like
The metric that matters
Replies per issue, and how many of those replies turn into calls
Realistic timeline
Twelve issues before you can judge your own list. Placements start landing around week eight.
A newsletter is two different channels sharing one word
Running your own list and getting placed in someone else's are not variations of the same activity. They fail for different reasons, they pay off on different timelines, and they need different work.
Your own newsletter is a writing and shipping problem. The hard part is cadence, and cadence is boring rather than difficult. It compounds over a year into an audience that no algorithm change can take from you.
Placement in other people's newsletters is a relationship problem, much closer to creator work. It is faster, it borrows someone else's trust, and it depends entirely on whether a writer wants to be associated with you.
| Your own list | Someone else's list | |
|---|---|---|
| Main constraint | Shipping every week | A writer saying yes |
| First results | Around issue twelve | Six to ten weeks |
| What automates well | Drafting, formatting, sending | Finding the right newsletters |
| What does not | Replies, the decision to cut an issue | The pitch, the terms, the follow-up |
| Owned by you | Entirely | Not at all |
Most companies should run both. They feed each other, because a placement sends people to a list that then has to be worth staying on.
What we do on your own list
The AI layer drafts. It has your product documents, your past writing, and whatever you said on a call this week, and it produces an issue that is roughly eighty percent there.
An operator does the rest, which is mostly subtraction. Cutting the paragraph that exists because you were proud of it. Rewriting the opening so it does not read like a company announcement. Deciding that this week's issue is thin and either finding the real story or shipping short rather than padding.
Then the send goes out on the same day, every week. That sounds trivial and it is the thing that actually fails. Weekly newsletters die in the third month, when the founder is busy and there is no external pressure to publish.
The replies are the channel
A newsletter that nobody replies to is a broadcast. The pipeline comes from the person who hits reply to say "we have this exact problem, how does it handle X".
Operators answer every one of those, personally, within the day, and they escalate the ones that are clearly a sales conversation to you rather than trying to run it themselves. They also read the replies for what they say about your positioning, because the way subscribers describe your product back to you is usually better copy than anything you wrote.
Getting into other people's newsletters
This is the half most founders skip, and it is the faster half.
The AI layer finds the newsletters whose readers are your buyers, which is a narrower list than it looks, then reads the archives so the pitch can reference something specific. An operator sends the note as a person and answers whatever comes back.
What follows is a negotiation. A writer might want payment, or a swap, or a piece written for them, or nothing at all because they liked the product. Terms differ every time. Someone has to agree what the copy will say, when it runs, and what happens if it slips, then confirm it in writing and follow up when it goes quiet.
No sequence tool does that part, for the same reason it cannot run a creator partnership. The other party is a person with their own priorities.
What we will not do
We will not buy a list, import contacts who did not subscribe, or send from a domain we control instead of yours. Deliverability is a reputation you build once and lose permanently, and the whole value of this channel is that it is yours.
Newsletter questions
Should a startup run its own newsletter at all?
Can AI write the newsletter?
How do you get into other people's newsletters?
Which email platform do you use?
Why do you report replies instead of open rates?
See what we would run on Newsletter.
Thirty minutes. You leave with the thirty-day plan either way.
30 minutes. You leave with the plan either way.