The short version: A good newsletter reads like a text from a smart friend, not a press release, it says one true thing well instead of six things badly, and it should make some people unsubscribe every single time you send it. If your newsletter never annoys anyone, it isn’t saying anything worth reading.
The thing nobody tells you about newsletter design
I spent about eighteen months of my early newsletter life obsessing over design. Header images, three-column layouts, a little icon next to every section title. I paid a designer good money to make my newsletter “look professional.”
Open rate barely moved. Click rate got worse, if anything, because all that design pushed the actual point further down the page.
Here’s what I’ve learned since, after sending a weekly email for the best part of a decade through the mess of the last five years rebuilding this business in public: the newsletters people quote back to me, forward to their team, reply to at midnight, are the plainest ones I write. Black text. One image at most. A subject line that sounds like a person wrote it.
What a good newsletter contains, in order
Strip away the branding and every newsletter that keeps people opening it, week after week, has the same skeleton. Mine does. So does Ann Handley’s, so does Sahil Bloom’s, so does the one from my accountant’s firm that I somehow never skip.
- A subject line that promises one specific thing. Not “This Week’s Update.” Something like “I lost a £4,000 client because I ignored this.”
- A first line that either continues the subject line or contradicts it. The reader decides in that first sentence whether to keep reading, on their phone, in about two seconds.
- One idea, not five. Most newsletters try to be a magazine. They have a “news roundup,” a “tip of the week,” a “quote I liked,” a “tool I found.” Readers skim all of it and remember none of it.
- A concrete example or number in the first third. Not “many businesses struggle with this.” A name, a figure, a date, a real thing that happened.
- A turn, where you say the uncomfortable bit. The sentence that makes the reader go “oh, that’s true and nobody says it.”
- One clear next step. Read this, try this, reply and tell me this. Not four links to click.
- A sign off that sounds like a human being closed their laptop, not a brand closed a campaign.
That’s it. Seven parts. Most newsletters have three of them and pad the rest with filler because someone told them “consistency matters more than quality,” which is true right up until the content is so thin that consistency just means being reliably boring.
A real one, because vague advice is useless
A few months back I sent a newsletter titled “The client who fired me and was right to.” No hedging, no clever wordplay, just the truth of what happened.
The story was short. A client I’d worked with for two years dropped me because I’d started treating the account on autopilot, sending the same monthly report format I’d used for every other client, not looking at what had changed in their business. She said, in the exit call, “I feel like I’m getting a template, not you.” She was right. I hadn’t updated my process for her in eight months.
That newsletter got a 61% open rate against my usual 40 to 44%, and 38 replies, which for a list of around 11,000 is a huge number for me. Not because the writing was clever. Because it was true, specific, and slightly embarrassing to admit. People don’t reply to “5 Tips for Better Client Retention.” They reply to a story where someone got it wrong and told you honestly.
That’s the pattern across every high-performing email I’ve ever sent, and I’ve now sent well over 400 of them: the ones with a real name, a real number, a real mistake, beat the ones with generic advice every single time. It isn’t close.
The uncomfortable bit most newsletter guides skip
Here’s the part that makes newsletter coaches uncomfortable, so they leave it out of their courses: a good newsletter should shrink your list.
Not by much. But if nobody ever unsubscribes, you’re writing for the middle of the room, saying things specific enough to be true but soft enough that nobody disagrees. That’s not a newsletter, that’s a company memo with a nicer font.
When I wrote about how most “AI marketing consultants” are just people who did a weekend course and slapped the word “AI” on their LinkedIn headline, I lost 43 subscribers in one day. Some of them were, unsurprisingly, people selling exactly that service. That newsletter also brought in three new client calls that week and got shared into two Slack groups I don’t have access to, because someone forwarded it. You cannot get the second thing without risking the first. If you want a list where nobody is ever mildly annoyed, you have to write nothing worth being annoyed by.
Length: the debate that doesn’t matter as much as people think
People ask me constantly whether a newsletter should be 300 words or 1,500. Wrong question. The right question is whether every sentence is doing work.
My shortest good newsletter was 210 words, a single story about a call gone wrong. My longest good one was 2,100 words, breaking down exactly how a fractional marketing setup works for a business that can’t afford a full CMO, which is the kind of thing I now write about when I cover what a fractional CMO for the AI era does day to day. Both performed well. Both had zero filler paragraphs. The badly performing ones in between, the 800-word ones that ramble through three half-finished thoughts, are the ones where I was writing to hit a schedule instead of writing because I had something to say.
If you have nothing worth 800 words this week, send 200 words. Nobody has ever unsubscribed from a newsletter for being too short and too honest.
Frequency and the myth of “consistency above all”
I send weekly, on a Thursday, because that’s what fits my working week. Weekly is not magic. What matters more than the day is whether you can sustain the standard you’re setting.
I’ve watched people I like launch daily newsletters with huge energy and burn out inside ten weeks, because daily quality writing at a professional standard is a full time job on its own, and most people already have one of those. If you’re building this alone, and I mostly have been for the last five years, weekly or fortnightly with real substance beats daily with padding, every time.
A concrete benchmark: if you can’t write your next issue in under two hours once you’ve had the idea, either the idea is too big for one email, or you’re trying to say too many things. Split it. Cut it.
The formatting rules that change results
- Short paragraphs, one to three sentences. On a phone, a six-line paragraph looks like homework.
- One link maximum inside the body, sometimes zero. Every extra link is a decision point that pulls the reader away from finishing your point.
- No stock photography. A screenshot of a real message, a real graph, a real thing, always beats a smiling stock model at a laptop.
- Bold sparingly, for the one sentence you’d want someone to screenshot and post on LinkedIn.
- A P.S. that carries actual weight. Some of my best replies come from the P.S. line, not the main body, because it reads like the thing you say as you’re walking someone to the door.
Where the writing itself needs to come from
A newsletter falls apart the moment it stops sounding like one specific person and starts sounding like “content.” I see this constantly in people who’ve come from a background of ghostwriting or agency work, where the goal was always to sound polished and neutral enough to represent any brand. If you’ve spent years chasing freelance writer vacancies that pay by the word for other people’s voices, your own newsletter is where you have to unlearn that habit and let the rough edges show. I wrote a longer piece on what that shift feels like in practice, including the parts of the freelance writer job that nobody puts in the ad, and the same lesson applies directly to newsletters: neutral, safe writing is the fastest way to be ignored.
The review and recommendation style newsletters have their own version of this problem, where every product mention reads like it came from the same press release. I built a specific structure for avoiding that when I wrote about how to write product reviews that convert, and it applies just as well to a “tool I liked this week” section in a newsletter: say what you used it for, say the one thing that annoyed you about it, then say whether you’d pay for it again. That structure alone will outperform a glowing generic recommendation nine times out of ten.
What to cut before you hit send
My personal checklist, used every single week before an email goes out:
- Cut the first paragraph if it’s throat clearing. Nine times out of ten, the real opening line is paragraph two.
- Cut any sentence that starts with “in today’s fast-paced world” or its cousins. If you wrote that, you’re stalling.
- Cut any claim you haven’t personally seen, tested, or lived. If you’re quoting a stat you found in someone else’s newsletter, either verify it or drop it.
- Cut the second call to action. One ask per email, always.
- Read it out loud once. If you stumble on a sentence, your reader will too.
A note on AI writing tools
People ask me whether it’s fine to draft a newsletter with AI now. It’s fine to draft with it. It is not fine to send what it hands you. AI is very good at producing the version of your newsletter that sounds like everyone else’s newsletter, because that’s what it’s trained on, the median of the internet. The story about the client who fired me, the specific £4,000 figure, the 43 unsubscribes, none of that comes from a tool. It comes from opening your own inbox and being honest about what happened in it that week. I’ve written before about how to spot the difference between someone using AI and someone hiding behind it, and the same test works for newsletters: does this sound like it could only have come from you, or could it have come from anyone with the same prompt?
The one metric worth watching
Forget vanity subscriber counts. The one number I track week to week is reply rate, because it tells me whether people are treating my newsletter as a conversation or as noise. A list of 3,000 people who reply is worth more than a list of 30,000 who don’t, both for revenue and for sanity. According to Forbes’ own reporting on newsletter economics, the smaller, highly engaged lists are increasingly what’s driving real business results, not sheer size, and that matches exactly what I’ve seen with my own numbers over the past two years.
Frequently asked questions
How long should a good newsletter be?
As long as it takes to say one true, specific thing well, no longer. That’s usually 300 to 900 words. Length should follow how much you have to say that week, not a template.
How often should I send a newsletter?
Weekly or fortnightly is sustainable for most solo writers and business owners. Daily sounds impressive but usually collapses into filler within a few months because few people can sustain original thinking at that pace alongside a real job.
What’s the biggest mistake people make with newsletter design?
Over-designing it. Heavy templates, multiple sections, and stock images push the actual point further down the page and lower both open and click rates. Plain text with one clear idea consistently outperforms a magazine layout.
Should every newsletter have a call to action?
One, not several. Multiple calls to action split attention and reduce the response to all of them. Pick the single next step you most want the reader to take and cut the rest.
