
Open rates for marketing email have fallen from 20-30% a decade ago to 10-15% now for most industries. It is tempting to treat this as death of email marketing. It is more accurate to treat it as the death of mass email, and the rise of something more selective and therefore more effective.
The emails that still work are the ones that were never mass emails.
Why open rates fell
Three things happened.
First, spam became ubiquitous. Inboxes filled with marketing email that people didn’t ask for. Open rates got buried under volume.
Second, email clients got better at filtering. Gmail’s smart compose, folder sorting, and promotion tabs mean that legitimate email is now often delivered somewhere a user never sees it.
Third, mobile email became dominant. Mobile inboxes are smaller. Preview panes are tiny. Subject lines that worked on desktop do not work on mobile. Most email is deleted by thumbnail.
None of this means email is dead. It means that low-effort email is now worthless.
The types of email that still work
Transactional. Receipts, password resets, order updates, delivery notifications. Open rates are 40-60%+ because these emails deliver something the person explicitly requested.
Highly segmented. Email sent to a very specific group with a specific need. A message about a new feature to existing customers who use that area of the product. A message about a sale on a specific product to people who have looked at it. Open rates are often 30%+ because relevance is high.
From a known sender about a known topic. A newsletter from someone you follow about their expertise. A message from a brand you care about. Open rates might be 15-25% because the audience chose to receive it.
Low frequency. A weekly update rather than daily. A monthly digest rather than every sale. Frequency and fatigue are intertwined. People have tolerance for a certain amount of email from a sender. Exceed it and they unsubscribe.
What does not work anymore
Blast email — the same message to everyone. Open rates are probably single-digit. Unsubscribe rates are measurable.
High frequency — multiple emails per day. Unsubscribe rates spike sharply above one email per week for non-transactional email.
Clickbait subject lines — “This one weird trick changed my life.” Gmail filters these now. If they get through, the person who was misled is more likely to unsubscribe than engage.
No unsubscribe path — this is now illegal in most jurisdictions and produces spam complaints that destroy your sending reputation.
The strategy that works
Build a list of people who want to hear from you. This is more work than buying a list or scraping email addresses. It is also 10x more effective. Offer something in exchange for an email address: a guide, a sample, early access.
Segment ruthlessly. Your customers are not a monolith. Some are using one feature, others are using something else. Some are new, others are long-time customers. Some have paid, others are free users. Send different messages to different groups.
Send email that delivers value. Not a “hello we exist” email, but an email that solves a problem, teaches something, offers a deal the recipient actually cares about, or provides an update on something they care about.
Match frequency to type. Transactional email can come every time something happens. Promotional email is usually one per week maximum. High-value updates might be monthly.
Track the right metrics. Open rate is no longer the metric that matters. Clicks matter. Conversions matter. Unsubscribe rate matters. An email with a 5% open rate but a 30% click rate from the people who did open it is successful. An email with a 20% open rate and zero clicks is a waste.
Test and refactor. Try different send times, subject lines, send frequency. The first version is probably not the best version.
The segmentation patterns that work
Behavioral. User has taken action X — purchased this product, viewed this page, downloaded this guide. Send them email relevant to X.
Demographic. If you know something about the person — they are in industry Y, at company size Z, in region W — send them relevant email.
Lifecycle. New users get onboarding emails. Long-time inactive users get win-back emails. Customers at risk of churn get retention offers.
Content preference. Ask people what they want to hear about and then send them that. Seems obvious, but most email programs do not do this.
Engagement tier. High-engagement subscribers (people who open and click most emails) can receive more frequent email. Low-engagement subscribers should receive less frequent email or be more targeted.
Fixing a declining email program
If your open rates are falling, the issue is usually not email itself. It is either poor targeting, poor content, or too much frequency.
Check these in order:
Deliverability. Are emails landing in inboxes or in spam? Check your reputation using tools like Google Postmaster Tools. If you are in spam, fix list quality and authentication (SPF, DKIM, DMARC) before doing anything else.
Segment. Are you sending the same email to everyone? If yes, segment and resend relevant messages. Do not expect everyone to be interested in everything.
Frequency. How often are you sending? If it is more than once per week, try reducing it. Fewer emails at higher quality often outperform frequent emails at lower quality.
Content. Are the emails delivering actual value? Not a product pitch, but something the person asked for or needs. “5 tips to improve X” is valuable. “We have a new feature” is not (unless the person uses that feature).
Subject line. Is it clear what the email is about? Is it honest? Is it mobile-friendly (short enough to display completely)?
If you fix these four things in order, open rates often recover to 15-20%, and engagement usually improves beyond that.
The future of email
Email is not going away. What is going away is the notion that you can send the same message to a huge list and expect response. Email is becoming more selective, more personalised, and therefore more effective for the people running it well.
The companies winning at email now are doing the opposite of what works for social or search. They are being more selective, not less. They are being more personal, not more automated. They are sending less often, not more.
This is the opposite of growth hacking culture, which has produced false positives in every other channel. Email is where selectivity actually works.