Latest / The Email Marketing Podcast with Fexingo: Newsletters, Open Rates, and Subscriber Growth / Why Send-Time Optimization Is Overrated for Newsletters
Transcript
- Lucas: There's this piece of email-marketing advice that gets repeated so often it's basically gospel: send at the perfect time. Tuesday at 10 a.m. Thursday at 2 p.m. Saturday morning for lifestyle content. I've even seen people recommend different times for different time zones, segmenting by the hour. Luna: Right, and there's no shortage of tools that promise to optimize send times for you. Mailchimp has it, ConvertKit has it, dozens of third-party apps. But you're skeptical? Lucas: I'm more than skeptical. I think for the vast majority of newsletter senders, obsessing over send time is a distraction from things that actually matter. And I want to look at the data behind that claim. Luna: Alright, I'm ready. And if these conversations have made you rethink something in your own marketing, honestly, if today was worth a coffee to you, that's the link — buy me a coffee dot com slash fexingo. No pressure, just a way to keep this ad-free. Back to send times. Lucas: Exactly. So let's start with one of the most cited studies on this. In 2016, AWeber — the email service provider — analyzed billions of emails and found that send-time optimization lifted open rates by only about 4 percent on average. Luna: Four percent. That's within the noise range for most A/B tests. You could get that bump just by changing your subject line from 'Newsletter' to 'This week's highlights.' Lucas: Precisely. And that 4 percent was the average across all senders. When they sliced the data by list size, the picture got even clearer. Lists under 10,000 subscribers saw no statistically significant lift at all. The tool just couldn't find a signal. Luna: So the smaller your list, the more the data is just random variation. Which is most newsletter creators, honestly. Lucas: Most of them. Now, the tools work by analyzing when individual subscribers have opened in the past and scheduling each email for that subscriber's peak time. It sounds smart. But there's a problem: open rates are a proxy, not a goal. Luna: Right. You want engagement — clicks, replies, conversions. An open at 3 a.m. when someone is just clearing their inbox might mean nothing. Lucas: Yes. And the other issue is recency. Most send-time optimization models rely on historical open data. But subscriber behavior changes. They get a new job, switch time zones, change their phone habits. The model is always a little behind. Luna: I've experienced that firsthand. I ran a B2B newsletter for about two years, and we tested morning versus evening sends. Our audience was mostly consultants and freelancers. Expected morning to win. Lucas: And? Luna: Evening — around 7 p.m. — had a 12 percent higher click-through rate. Our theory was that they were too busy with client work during the day and checked personal email at night. But we never would have guessed that without testing. Lucas: That's the key: testing. Not relying on a tool that claims to know your audience better than you do. If you test two or three time slots over a few weeks, you'll learn more than any algorithm can tell you. Luna: And you'll have a rationale for the choice. 'We send at 7 p.m. because our audience is consultant-heavy and they check email after client hours.' That's a story that makes sense. Lucas: Exactly. Compare that to 'The algorithm says you should receive this now.' It's opaque. It doesn't build trust. Luna: So what's your practical advice for someone listening who's using a send-time optimization tool right now? Lucas: Turn it off. Or at least pause it for a month. Pick one consistent send time — say Tuesday at 10 a.m. — and just send. Track your open and click rates for four weeks. Then switch to a different time for another four weeks. Do that a couple times. Luna: And if you see no difference, pick the time that's easiest for you to produce consistently. Lucas: Exactly. Consistency is actually more important than the specific hour. If subscribers know your newsletter arrives every Tuesday morning, they start to expect it. That expectation itself lifts engagement. Luna: There's also the sender reputation angle. If you send at wildly different times for different segments, your sending pattern looks erratic to mailbox providers like Gmail. They might flag you as less predictable. Lucas: That's a great point. Inbox placement is a whole other topic, but yes — consistent volume and timing help build a positive sending reputation. So send-time optimization could actually hurt deliverability if it makes your sending pattern look spammy. Luna: So the two things that matter more: content quality and sender reputation. Not whether you hit 10 a.m. or 2 p.m. Lucas: Right. And I'd add a third: list hygiene. If you're sending to stale addresses, no send time will save you. But that's a previous episode. Luna: Episode 4, actually. List churn is healthy. Lucas: Good memory. So the bottom line: if you can articulate why a specific send time works for your specific audience — like Luna with her consultant list — great. Use it. But if you're just following 'best practices' or trusting a black-box tool, you're probably wasting energy. Luna: One more thing: what about time zones? Should you send at 10 a.m. in your own time zone, or should you try to send at 10 a.m. for each subscriber's time zone? Lucas: That's actually a different feature, often called 'time zone sending.' Some research suggests it can lift open rates by a few percentage points. But again, the effect is small. And it requires more complex infrastructure. For most small to mid-size lists, I'd say pick one time zone — preferably the one where most of your subscribers are — and send there. Luna: And if your audience is global, you could rotate send times or send on different days for different regions. But that's a lot of work for marginal gain. Lucas: It is. And the data from our own Fexingo network shows that our highest-open newsletters are the ones sent at the same time every week, regardless of what that clock says. Luna: Consistency over optimization. I like that as a takeaway. Lucas: One final thought: the tools are getting better. Eventually, machine learning might actually deliver meaningful lifts. But right now, in May 2026, the evidence is still thin. Don't pay for a feature you can replicate with a simple A/B test and a calendar. Luna: And if you do test something, let us know. We might feature your results in a future episode. Lucas: Send us a note. That's it for today. Next time, we're going to talk about the one metric you should never optimize for: the unsubscribe rate. Luna: Intriguing. See you then.