I have sent somewhere north of three hundred newsletter issues over the past six years. Two of those newsletters died quietly. One of them still pays part of my rent. The difference between them was never the tool, the template, or the send time. It was whether I had figured out what the reader wanted before I opened the editor.
That sounds obvious written down. It was not obvious to me at the start. My first newsletter was a weekly roundup of "things I found interesting," which is a polite way of saying it had no point. Open rates slid from 48% on issue one to 19% by issue twenty, and I blamed Gmail's Promotions tab instead of the real problem: nobody knew what they were going to get, so nobody had a reason to open it.
This guide is everything I wish someone had handed me before that first send: the real numbers for 2026, how to give your newsletter a job, subject lines that do not oversell, structure for skimmers, and the technical setup that decides whether your email reaches the inbox at all.
Why most newsletters get ignored
Before fixing anything, it helps to name the three failure modes I see over and over.
No clear promise The reader cannot predict what an issue contains. Every open is a gamble, and gamblers stop showing up. | Written for the sender, not the reader Company news, feature announcements, and "we are excited to share" openers. The reader has no stake in any of it. | Never lands in the inbox Authentication is missing or broken, the list is full of dead addresses, and the email goes to spam before anyone can judge the writing. |
What "good" looks like in 2026
Newsletter benchmarks are unusually messy right now, and the reason is Apple. Mail Privacy Protection, launched with iOS 15 in September 2021, pre-downloads email content, including the tracking pixel that registers an open, for Apple Mail users with the feature enabled. That counts as an open whether or not a human ever looked at the message.
This is why different reports disagree so wildly. Brevo's 2026 benchmark, built on data from more than 175,000 accounts, reports a 20.73% "standard" open rate that filters out Apple's automated fetches, and a 33.87% rate that includes them. MailerLite, which does not strip them out, reports a 43.46% average open rate across 3.6 million campaigns sent in 2025.
| Metric | Figure | Source | Note |
|---|---|---|---|
| Open rate (MPP filtered) | 20.73% | Brevo 2026 | Closer to real human opens |
| Open rate (MPP included) | 33.87% | Brevo 2026 | What most dashboards show |
| Open rate (all campaigns) | 43.46% | MailerLite 2025 data | Includes Apple pre-fetches |
| Newsletter open rate | 41.24% | beehiiv State of Newsletters | Newsletter platforms specifically |
| Click rate | 2.09% | MailerLite 2025 data | Range across industries: 0.83% to 4.90% |
| Click-to-open rate | 6.81% | MailerLite 2025 data | Up from 5.63% in 2024 |
| Click rate (marketing emails) | 2.27% | Brevo 2026 | Newsletters, promos, announcements |
The practical takeaway: treat open rate as a rough directional signal and put your attention on click rate, reply rate, and unsubscribe rate, which require a human to do something on purpose. If your newsletter is built to be read inside the inbox with few links, track replies and retention as well.
Here are those simple 8 Steps
STEP 1 - Give the newsletter one job
The single most useful exercise I ever did was writing one sentence: "Every Tuesday, this newsletter tells busy product managers which three things changed in analytics tooling and why it matters." That sentence killed 60% of the ideas I had been cramming into each issue, and the open rate on the survivor climbed back over 40% within two months.
Pick a format and commit to it. These three hold up best in my experience.
| Format | What it is | Works best when | Watch out for |
|---|---|---|---|
| Curated digest | A short list of links with your commentary on each | Your audience is time-poor and trusts your taste | Commentary must add something; a bare link list gets skipped |
| Single-idea essay | One argument, one story, or one lesson per issue | You have a distinct point of view | Needs a strong hook in the first two lines or people bail |
| Practical playbook | One actionable how-to per issue | Readers came for skills, not entertainment | Keep the scope narrow enough to finish in one sitting |
Hybrids can work, but they still need a stable shape. Readers should be able to describe your newsletter to a friend in one sentence. If they cannot, you have not finished defining it.
STEP 2 - Subject lines and preview text
The subject line has one job: get the right people to open. Not everyone, the right people. Clickbait can spike a single issue and then wreck the next five, because readers learn that your subject lines lie.
Rules I actually follow
• Say what is inside. Specific beats clever nine times out of ten.
• Keep the important words in the first 40 characters, since mobile clients truncate.
• Do not fake urgency. If it is not time sensitive, do not pretend it is.
• Use the preview text as a second subject line, not as a repeat of the first.
• Avoid ALL CAPS, strings of punctuation, and words like "free" in isolation. They will not automatically trigger spam filters, but they read like spam to humans.
Before and after examples
| Weak subject line | Stronger subject line | Why it works |
|---|---|---|
| Newsletter #47 | The pricing page mistake I made twice | Specific, first person, implies a lesson |
| Big news! | We are shutting down the free plan (here is why) | Tells the reader what is happening |
| Don't miss this | How a 2-line email won back 412 churned users | Concrete result the reader might want |
Test subject lines, but test them properly. An A/B split on 200 subscribers proves nothing. Use it on lists of at least a few thousand and judge the winner on click rate, not just opens.
STEP 3 - Structure the body for skimmers
Most readers will not read your newsletter top to bottom. They will scan it, decide in a few seconds whether it deserves attention, and then either slow down or delete. Write for the scan first and the deep read second.
The structure I use for almost every issue 1. Opening line: the single most interesting thing in the issue, stated plainly. No "Hope you are well." No weather. 2. Context: two or three sentences on why it matters right now. 3. The body: broken into clearly labelled sections, each one scannable on its own. 4. One call to action: reply, click, or forward. Pick one. 5. Sign-off: a real name, a sentence about what is coming next issue. |
Some formatting habits that measurably improved my read time:
• Bold one phrase per section, the one that carries the point. Not full sentences.
• Keep paragraphs to three lines on mobile. Long blocks get skipped wholesale.
• Use subheads that say something. "Why the cheaper plan cost us more" makes people stop. "Section 2" does not.
• Limit links. Three well-chosen links get more total clicks than twelve scattered ones.
On length: there is no correct word count. A digest can run 300 words and an essay can run 1,500. What matters is that the length matches the promise you made on the signup page.
STEP 4 - Write like one person talking to one person
The newsletters I keep opening year after year share a voice. They sound like a specific human, with opinions, occasionally wrong, always direct. The ones I unsubscribe from sound like a committee.
| Do this | Not this |
|---|---|
| "I tested this for three weeks and it broke twice." | "Users may experience occasional instability." |
| "Here is what I would do in your position." | "Best practices suggest the following approach." |
| "Reply and tell me if this landed." | "Feel free to reach out with any feedback." |
| Short, plain words | Corporate phrasing and stacked adjectives |
Write each draft as an email to one specific subscriber you can picture, then read it aloud. Anything you would not say to that person gets cut. And search your last five issues for "we": if it outnumbers "you" by a wide margin, the newsletter is about the sender.
STEP 5 - Cadence and consistency
Readers form habits around predictable sends. The exact day and time matter less than the fact that it is always the same day and time.
| Cadence | Best suited to | Risk |
|---|---|---|
| Daily | News, markets, very short digests | Burnout for the writer, fatigue for the reader |
| Weekly | Most newsletters; the default I recommend | Missing a week is very visible |
| Biweekly | Deep essays, small teams | Easy to lose momentum and drop to monthly |
| Monthly | Company updates, low-volume niches | Readers forget who you are between sends |
I have run weekly for five years. When I slipped to "whenever I have something" for a stretch in 2023, open rates dropped 11 points in eight weeks and took four months to recover. A shorter issue on schedule beats a better issue that is late. If you cannot sustain weekly, choose biweekly and say so on your signup page.
STEP 6 - The deliverability basics that quietly kill newsletters
This is the section most writing guides skip, and it is the one that most often explains a sudden drop in opens. Since February 2024, Google and Yahoo have enforced requirements for anyone sending more than 5,000 messages a day to their users, and Microsoft announced matching requirements for Outlook in 2025. Even if you are well under that volume, treating these as mandatory is the safest path.
| Deliverability checklist | |
| ☐ | SPF and DKIM are set up on your sending domain and passing |
| ☐ | DMARC record exists (a policy of p=none is the minimum to start) |
| ☐ | One-click unsubscribe headers are present (most reputable platforms add these automatically) |
| ☐ | Unsubscribe requests are honored within two days |
| ☐ | Spam complaint rate stays under 0.3%, and ideally under 0.1%, in Google Postmaster Tools |
| ☐ | Sending domain matches the from address rather than a free webmail address |
| ☐ | Bounces are cleaned automatically so hard-bounced addresses never get a second send |
| ☐ | Inactive subscribers get a re-engagement send and are then suppressed if they stay silent |
Google publishes the full list in its Email sender guidelines, and it is worth reading directly rather than relying on summaries, because the details on DMARC alignment and spam thresholds are where people trip up.
Two things I learned the hard way: warm up a new sending domain gradually rather than blasting a large list on day one, and stop emailing people who never open. A list of 4,000 engaged readers beats a list of 20,000 with 15,000 dead addresses, both in results and in inbox placement.
STEP 7 - Grow with readers who will actually stay
Where a subscriber comes from predicts how they behave. Readers who arrive through a recommendation or a piece of your writing open at far higher rates than readers acquired through giveaways, unrelated lead magnets, or purchased lists (never use those).
Growth sources ranked by subscriber quality, in my experience
1. Forwards and word of mouth. Ask readers to forward the issue to one specific person who would find it useful.
2. A signup page with a sample issue. Show people exactly what they will get.
1. Cross-promotion with adjacent newsletters. Swap mentions with writers whose audience overlaps but does not duplicate yours.
2. Your own site and social channels, linking to the signup page with a clear promise.
3. Paid acquisition. Works at scale but produces the coldest subscribers; measure retention, not signups.
Whatever the source, send a welcome email immediately. Klaviyo's 2026 benchmark report found automated flows generating around three times the click rate of regular campaigns. Restate the promise, set the cadence expectation, and give the reader one thing worth reading right now.
STEP 8 - Measure what matters
Here is the dashboard I check after every send, and what each number tells me.
| Metric | What it tells you | What I do with it |
|---|---|---|
| Click rate | Whether the content prompted action | Compare against my own trailing average, not industry averages |
| Click-to-open rate | Whether the body delivered on the subject line | If opens are high and CTOR is low, the subject line overpromised |
| Reply rate | Whether readers feel a personal connection | The number I care about most for a written newsletter |
| Unsubscribe rate | Whether an issue broke the promise | A spike after one issue is a signal to review that issue specifically |
| Spam complaints | Deliverability risk | Anything trending toward 0.1% gets immediate attention |
| Subscriber retention (90 day) | Whether new readers stick | The real growth metric; signups without retention are noise |
Track your own trendline rather than industry averages, and segment before judging: your most engaged 20% might open at 60% while a cold segment sits at 12%, and the blended number hides both.
Mistakes that keep showing up
| Mistake | Why it hurts | Fix |
|---|---|---|
| Opening with "Hope this finds you well" | Wastes the most valuable line in the email | Lead with the most interesting thing in the issue |
| Five calls to action | Readers do nothing when asked to do everything | One primary ask per issue |
| Image-only emails | Blocked images mean a blank email; screen readers get nothing | Text first, images as support, alt text always |
| Never asking for replies | You lose the fastest feedback channel there is | End with a direct question at least every few issues |
| Buying or scraping lists | Wrecks deliverability and violates consent laws in most regions | Never; grow slowly instead |
Pre-send checklist
I run through this before every issue. It takes about four minutes and has saved me from several embarrassing sends.
| Before you hit send | |
| ☐ | Subject line says what is inside, no overpromise |
| ☐ | Preview text is written, not defaulted to the first line |
| ☐ | First sentence is the strongest sentence |
| ☐ | One clear call to action |
| ☐ | Every link tested, including the unsubscribe link |
| ☐ | Read aloud once for voice |
| ☐ | Checked on a phone, not just desktop |
| ☐ | Reply-to address is a real person or monitored inbox |
| ☐ | Sent a test to yourself and read it as a subscriber would |
Final verdict
After six years and several hundred issues, the thing I keep coming back to is that a newsletter is a relationship, not a channel. Every issue either adds trust or spends it. The technical side matters, and I have watched broken authentication erase months of good writing, but once your email reliably lands in the inbox, the only question that matters is whether the reader is glad they opened it.
If I had to compress this whole guide into three instructions, they would be these.
1. Make one clear promise and keep it every single issue. 2. Write like one person to one person, with the strongest line at the top. 3. Measure clicks, replies, and retention instead of chasing an open rate that Apple has been inflating since 2021. |
The newsletter that still pays part of my rent is not clever. It goes out at the same time every Tuesday, it says what it will contain in the subject line, it is usually under 800 words, and it ends with a question. That consistency, more than any single brilliant issue, is what kept people reading. Start there, and adjust only once you have ten issues of your own data to look at.
