FAQ Friday: Can an AI Agent Run Your Newsletter? (Mine Wrote 22 Issues. The List Has 7 Subscribers.)

AI newsletter automation: a glass mail tray full of blank envelopes next to a small glass bowl holding only a few amber lights

This week’s question comes up every time I show someone the fleet: “Can an AI agent run my newsletter by itself? Write it, send it, every week?”

Short answer: the writing, yes. The sending, mostly. But the newsletter was never the hard part, and my own receipts prove it.

The short answer: it writes every week, and it mostly sends

Every Friday at 08:05 an agent in my fleet wakes up, reads a brief another agent prepared the day before, picks the week’s best post, writes the issue and hands it to Sendy to send. Nobody touches it. I pulled every run log this morning (20 November 2026):

  • 22 Fridays, 22 issues written. The first went out on 19 June. The agent has not missed a week.
  • 19 were sent. Every Friday from 19 June to 23 October.
  • The last 3 were not. 30 October, 6 November and 13 November are sitting in Sendy as drafts.

So yes, an agent can write a newsletter on schedule for five months straight. That part is solved.

Why did 3 issues never go out?

Because the list ID the agent was given stopped matching any list in Sendy. Ask Sendy for that list’s subscriber count today and it answers “List ID not passed”. The agent did the right thing: it didn’t guess another list, it saved each issue as a draft and reported fail.

Jon Jones

⚡ GET THE AI EDGE

Weekly AI tips that actually save you time and money. No fluff, no hype — just what works.

Newsletter Signup - Blog CTA

Here’s the catch. The job still exits cleanly every time, so anything that only checks “did it crash?” sees three green Fridays. It’s the same trap I wrote about in failing soft instead of loud. My nightly watchdog reads the agent’s own result line, not just the exit code, which is why I know. Fixing the list ID is on my list, not done. Today’s issue is due at 08:05, so unless that changes it will be the 4th draft.

What’s the real bottleneck?

This is the part nobody puts in the demo. That list has 7 active subscribers. It has had 7 the whole time.

  • My signup form has collected 5 entries since 12 June. The last one was 1 November.
  • My CRM holds 17 contacts. My setup notes say every signup gets a “newsletter” tag. That tag has 0 contacts. All 17 carry a different one.

So for five months an agent has written a weekly issue to roughly the number of people who fit around my dinner table. The writing scaled perfectly. The audience didn’t, because an agent writing emails does nothing to get people onto the list.

Do this today

  1. Check the list before the writer. Pull your active subscriber count and your signups for the last 30 days. If both are small, more issues won’t help. Point the effort at the signup form first.
  2. Make the send step prove it sent. Your agent should report the campaign status your email tool returns, not “done”. “Saved as draft” and “sent” must look different in your logs.
  3. Watch the result, not the exit code. A run that finishes cleanly and reports a failure should still wake you up. Count the fails per week.
  4. Test the plumbing monthly. List IDs, tags and API keys drift. A one-line count of each list is enough to catch it.

The takeaway: an AI agent can run your newsletter, but it can’t grow your list. Automate the writing, then put the hours you saved into the signup side.

For the wider picture of how email fits into a small business stack, here’s my take on email marketing and automation, and why I give my writer agents a “do not claim” line. If you want a newsletter pipeline that reports honestly when it breaks, book a strategy session.

The AI Playbook — Free Download

📥 FREE: THE AI PLAYBOOK

The exact tools and workflows I use to run a one-person agency. 25 years of marketing experience distilled into an actionable guide. Yours free.

Lead Magnet - AI Playbook

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *