How to Write a Weekly Newsletter That People Actually Open

Launch a weekly newsletter and the first six weeks will lie to you. Open rates sit at 45-55% because your list is small, warm, and mostly made up of people who signed up recently and remember why. Then week eight arrives, the number drops to 30%, week sixteen it’s at 22%, and the team starts debating subject lines as if that’s the whole problem. It rarely is. Subject lines affect the margin, but the structural habits underneath a newsletter — what it’s actually for, how it’s built, how the list is maintained — determine whether that decay flattens out at a respectable 35% or keeps sliding toward 10%.

Decide what job the newsletter is actually doing

Most newsletters fail because nobody picked a single job for them before launch, so every issue tries to do three things at once — promote the product, share company news, and provide genuine value — and readers can feel the confusion even if they can’t name it.

A newsletter that exists to nurture trial users toward conversion reads completely differently from one that exists to build a personal audience around a founder’s expertise, which reads differently again from one that exists to keep paying customers engaged so they don’t churn. Pick one primary job. The other things can happen occasionally, but the primary job should be identifiable from the first three issues a new subscriber receives, because that’s the window where they decide whether to keep opening.

A useful test: if you deleted your company’s name and logo from the last five issues, would a stranger still find them worth reading. If the honest answer is no — if the content only has value because of who’s promoting it — the newsletter is functioning as an ad delivery mechanism with a value-content veneer, and open rates will decay faster than a newsletter that would hold up on its own.

The structural habit that matters more than any single subject line

Consistency of format beats quality of any individual issue for long-term open rates, because familiarity is what makes a newsletter part of someone’s routine rather than one more thing competing for attention. A reader who knows an issue always opens with a specific short section, always has three curated links, always closes with the same sign-off, develops a habit around that structure the same way people develop habits around a favorite podcast’s intro music.

This means resisting the urge to reinvent the format every few issues to “keep it fresh.” The freshness should live in the content, not the structure. Morning Brew, one of the most studied newsletter case studies in the industry, has kept essentially the same section structure for years — the variation is entirely in what fills those sections each day, not in reworking the format itself.

Pick a consistent length and stick close to it. If issue one is 400 words and issue six is 1,800 words, subscribers can’t calibrate how much time to set aside, and that uncertainty itself suppresses opens over time — people are more likely to defer opening something whose time cost is unpredictable, and deferred opens frequently become un-opens.

Subject lines: what actually moves the number

Specificity beats cleverness almost every time. “3 things that broke our onboarding funnel last month” will outperform “Onboarding secrets revealed” in essentially every test, because the specific version signals exactly what’s inside and lets a reader self-select based on real interest, while the vague version relies purely on curiosity, which is a weaker and more fatigue-prone motivator over a long subscriber relationship.

Numbers and concrete nouns outperform abstract claims. “The pricing page change that added $40k in MRR” beats “How we improved our pricing” because it’s verifiable and specific rather than a vague promise of insight.

Avoid the words that trigger both reader fatigue and spam filters simultaneously — “free,” excessive punctuation, ALL CAPS words, and phrases like “don’t miss out” have been so overused in promotional email that they now suppress opens even among engaged subscribers, independent of any deliverability penalty.

Test preview text as seriously as the subject line itself. Most email clients show 40-90 characters of preview text right next to the subject, and a generic or missing preview (“View this email in your browser…”) wastes free real estate that could be doing persuasive work. Treat it as a second, shorter subject line that either reinforces or extends the promise of the first.

The list hygiene work nobody wants to do

The single biggest lever for maintaining open rates over time isn’t better content, it’s actively removing people who’ve stopped opening. This feels counterintuitive — the instinct is to protect list size — but a bloated list of disengaged subscribers actively damages deliverability for the whole list, because inbox providers use engagement signals to decide whether your emails land in the primary inbox or get filtered, and a large block of subscribers who never open drags that signal down for everyone.

Run a re-engagement sequence for anyone who hasn’t opened in 60-90 days: a short email acknowledging the silence, asking directly if they still want these emails, with a clear one-click way to confirm interest or unsubscribe. Whoever doesn’t respond gets removed. This will shrink the list, sometimes by 15-30%, and it will almost always raise the open rate on the remaining list immediately, both because the math changes and because deliverability genuinely improves once inbox providers see a cleaner engagement signal.

Segment repeat non-openers out of send frequency before they unsubscribe entirely. Someone who hasn’t opened in 45 days but hasn’t unsubscribed either is telling you something — dropping them to a monthly “best of” digest instead of the weekly send often keeps them nominally subscribed and occasionally re-engaged, rather than losing them to an unsubscribe or, worse, a spam complaint.

Send time and cadence matter more than most teams test for

Weekly cadence works well for most B2B newsletters because it matches how often a reader can reasonably expect meaningfully new content without feeling spammy, but the specific day and time within that week is worth testing rather than assuming. Tuesday through Thursday mornings in the recipient’s time zone tend to outperform Monday (inbox triage day, low attention) and Friday afternoon (checked out) for most B2B audiences, but this varies enough by audience that a genuine A/B test of send time on your own list beats copying an industry benchmark.

Once you find a time that works, protect it religiously. A newsletter that shows up reliably at 8am Thursday becomes part of a routine; one that arrives at unpredictable times loses the routine advantage entirely, even if the average send time is similar, because subscribers can’t build an expectation around it.

A Worked Example: What Decay Actually Looks Like Without Intervention

Take a newsletter that launches with 2,000 subscribers and a 50% open rate — 1,000 opens in week one. With no list hygiene and no structural discipline, a typical decay curve looks like this: 38% by week eight (760 opens on the same 2,000-person list, since growth has been slow), 26% by week sixteen, and 18% by week twenty-four — not because the content got worse, but because the list keeps accumulating people who signed up once, stopped caring, and never left. By week twenty-four, a meaningful share of that list — often 20-30% — hasn’t opened a single issue in the last two months, and that dead weight is actively suppressing deliverability for the remaining engaged readers, since inbox providers weight sending reputation against aggregate engagement, not against your best subscribers alone.

Run the 60-90 day re-engagement sweep described below at week twelve instead of waiting until week twenty-four, and the trajectory changes meaningfully: removing the roughly 400-500 subscribers who’ve gone dark drops total list size to around 1,600, but because those subscribers were dragging down both the open-rate denominator and the actual deliverability signal, the rate on the cleaned list often jumps back up to 32-35% immediately — not because anyone did anything differently, but because the math and the inbox-placement signal both improved simultaneously. Waiting until week twenty-four to run the same cleanup recovers less ground, because by then deliverability damage has had longer to compound and some subscribers who might have re-engaged at week twelve have fully forgotten why they subscribed at all.

The Failure Mode: Over-Rotating on Subject Line Testing While Ignoring the List

The most common misdiagnosis when open rates decline is assuming it’s a copywriting problem and pouring effort into subject line testing — running dozens of A/B variants, hiring a copywriter, restructuring the send calendar around “optimal” subject line formulas. This can produce a small, real lift (a few percentage points) but it’s working on the margin of a much bigger problem if list hygiene has been neglected for months, because no subject line can overcome a list that’s 30% dead weight actively harming deliverability.

The tell that you’re solving the wrong problem: if subject line tests show statistically meaningful differences between variants but the 90-day rolling average keeps declining regardless of which variant wins, the issue isn’t subject lines — it’s the underlying list health or a structural problem with format consistency. Check list hygiene and format consistency before investing further in subject line optimization; the order matters because a clean, engaged list makes every subsequent subject line test more sensitive and more trustworthy, while testing subject lines against a decaying list produces noisy results that are hard to act on with confidence.

Small Lists and B2C Consumer Lists Behave Differently

Everything above assumes a moderate-to-large B2B list where deliverability signals and list hygiene dominate the story. Two situations need adjustment. First, very small lists (under 500 subscribers) don’t have enough volume for most of the A/B testing described above to reach statistical significance in a reasonable timeframe — send time tests, subject line tests, and format tests all need a larger sample to trust the result, so a small list should prioritize the structural habits (consistent format, clear single job, list hygiene) over granular testing, since testing noise will otherwise be mistaken for signal.

Second, consumer newsletters with a broader, less professionally-motivated audience tend to have naturally lower baseline open rates than B2B lists (a B2C newsletter settling at 15-20% can be healthy, where the same number would signal a real problem for a B2B list), and they also see much stronger day-of-week effects tied to leisure time rather than work schedules — weekend and evening sends often outperform weekday mornings for consumer content, the reverse of the B2B pattern described above. Benchmark against your own list’s history and against genuinely comparable newsletters in your category, not against a generic industry average that doesn’t specify B2B versus B2C.

Building in a feedback loop instead of guessing

Most newsletter teams look at open rate and click rate and stop there, which tells you what happened but nothing about why. Add a lightweight, low-friction feedback mechanism every 8-10 issues — a single clickable question at the bottom (“more of this / less of this,” linked to two different landing pages that just log the click) gives you a running signal on content direction without requiring a full survey that most people will ignore.

Read reply emails personally, especially early on. A newsletter that allows replies (rather than sending from a no-reply address) generates a steady trickle of direct subscriber feedback that’s more honest and specific than almost any survey response, because people who bother to reply to an email are self-selecting as engaged enough to have an actual opinion worth reading.

Track open rate trend over rolling 90-day windows rather than issue-to-issue, since single-issue open rates swing based on subject line variance and shouldn’t be over-read. What matters is whether the 90-day rolling average is flat, declining, or improving — that’s the number that tells you whether the structural habits above are actually working, not any single week’s result.