Latest / The Newsletter Podcast with Fexingo: Substack, Beehiiv, and Email Newsletter Business / How Newsletter Writers Use Email List Segmentation to Double Revenue
Transcript
- Lucas: Let's talk about something almost every newsletter writer knows they should do but most put off — segmenting their email list. Not just free versus paid, but actually carving your audience into groups based on what they read, how often they open, and what they click. Luna: I'm guilty of that. I have a small list and I just send the same thing to everyone. It feels like segmentation is for the big players. Lucas: That's the perception, but the data says otherwise. I was looking at a case study from a mid-size business newsletter — call it 'Market Pulse' — they had about forty thousand subscribers. They moved from a single weekly send to three distinct email flows based on engagement. Within three months their premium conversion rate doubled. Lucas: And here's the kicker — they did it with a simple three-segment model: 'high-intent' subscribers who had clicked a link in the last thirty days, 'curious' subscribers who had opened but not clicked, and 'dormant' subscribers who hadn't opened in sixty days. Each group got a different treatment. Luna: So what did they send to each group? I imagine the dormant ones need a re-engagement sequence, not just the same newsletter. Lucas: Exactly. The high-intent group got an early access preview of their paid content with a direct call to action. The curious group got a 'here's what you missed' digest with a softer upsell. And the dormant group got a 'we miss you' email with a free gift — a downloadable PDF of their most popular posts. Luna: That last one is interesting because it's not asking for money, just rekindling the habit. How did they measure success? Lucas: They tracked three metrics: open rate for the dormant re-engagement, click-through rate for the curious digest, and conversion rate for the high-intent early access. The dormant re-engagement brought back about twelve percent of that segment. And the high-intent group converted at nearly eight percent — compared to their previous average of three percent. Luna: Eight percent from a single email? That's pretty impressive. I wonder if there's a risk of over-segmentation though — getting into such small groups that you're basically sending individual emails. Lucas: That's a real concern. The newsletter writer I interviewed for this — his name is Mark Chen, runs a marketing analysis newsletter — he told me his rule of thumb is to have at least one thousand subscribers per segment to maintain statistical significance. Otherwise you're guessing, not segmenting. Lucas: And he also said the biggest mistake is segmenting based on demographics alone — age, location, job title — because those don't predict reading behaviour nearly as well as engagement data. Luna: That makes sense. My subscribers who are CEOs don't all read the same way. Some open everything, some never open. So behaviour-based segments are more actionable. Lucas: Precisely. And the beauty is that the data is already there in your email platform. You don't need a data scientist. Most good newsletter tools — Beehiiv, ConvertKit, Substack — have built-in segmentation that lets you create rules based on open and click history. Lucas: One thing that came up in the interview that I thought was clever: Mark uses what he calls 'content preference segmentation'. When a new subscriber joins, he asks them to pick two or three topics from a list. Then he tags them and sends topic-specific emails. His open rates went from thirty-five to fifty-two percent on those targeted sends. Luna: That's a huge jump. But doesn't that add a lot of overhead? You're writing multiple versions of each newsletter. Lucas: He doesn't write multiple full newsletters. He writes one main piece and then swaps out a section — like a case study or a tool recommendation — based on the segment. So it's maybe an extra hour per week. He's a solo writer. Luna: Alright, I'm sold. But I want to talk about the ethical side. When you start tracking clicks and opens to segment people, where's the line before it feels intrusive? Lucas: That's a fair question. I think the key is transparency and value. If the subscriber knows you're using their behaviour to send them more relevant content — and they can see the benefit — it's usually fine. The problem is when you start using it to manipulate or to sell without consent. Lucas: Mark actually includes a line in his welcome email: 'I use your reading habits to send you the most relevant content. You can opt out of personalization anytime.' He said only about two percent opt out. Luna: That's reassuring. And honestly, if I'm getting emails that actually interest me, I'm not going to complain about the tracking. So what's the first step for a newsletter writer who wants to start segmenting but feels overwhelmed? Lucas: Start with one segment. Pick your most engaged subscribers — anyone who clicked a link in the last thirty days — and send them something slightly different. It could be as simple as a different subject line or a bonus link. See if that group responds better. Then build from there. Luna: That sounds manageable. I could probably do that this week. And if these marketing conversations have sparked something you've actually used in your own work, supporting the show is a nice way to keep it going. Lucas: Yeah, listener support is what keeps this ad-free. If you've gotten value from the episodes, you can contribute at buy me a coffee dot com slash fexingo. It's a small gesture but it really helps us keep doing deep dives like this. Luna: Agreed. And now, back to segmentation. Lucas, you mentioned that Mark uses a 'dormant' segment. What's the timeline for that? Sixty days seems short. Lucas: It varies. For a daily newsletter, sixty days is already generous. For a weekly, maybe ninety. The idea is to catch people before they've completely checked out. If they haven't opened in sixty days, the chance they'll spontaneously open the next one is low. So you intervene with a re-engagement email. Luna: What if they still don't open after that? Do you eventually remove them? Lucas: Mark's policy is three strikes. After the re-engagement email, if they don't open in another thirty days, he sends a final 'should I remove you?' email. If no response, he unsubscribes them. His list shrank by about ten percent, but engagement metrics went up across the board. Luna: That's actually a good practice for deliverability too. Internet service providers look at engagement rates. A clean list means better inbox placement. Lucas: Exactly. And that's the paradox — by removing people, you actually get more opens and clicks, which makes the platform favour your emails. Segmenting helps you identify who to keep and who to let go. Luna: I also wonder about the tools. You mentioned Beehiiv and ConvertKit. Do they have segmentation built-in, or do you need a separate CRM? Lucas: Most modern email service providers have decent segmentation. Beehiiv has a 'segments' feature where you can create rules based on any behaviour. ConvertKit has tags and segments. Substack is more limited — you can segment by paid versus free, but not by behaviour natively. You'd need to manually export and import. Luna: So if you're on Substack and want to do this, you might need to switch platforms or use a workaround. Lucas: Or use a third-party tool like Zapier to sync Substack subscriber data into a separate email tool. But that's extra complexity. Most serious newsletter writers I know who do heavy segmentation are on Beehiiv or ConvertKit. Luna: One last thing — how do you measure the ROI of segmentation? Like, if I spend an extra hour a week on it, how do I know it's worth it? Lucas: Track your conversion rate and your average revenue per subscriber. Mark saw his average revenue per subscriber go from $0.42 to $0.89 in three months. That's a $0.47 increase times forty thousand subscribers — roughly eighteen thousand dollars more per month. For an extra hour a week, that's a pretty good return. Luna: That's compelling. I think the takeaway is: start small, pick one segment, and iterate. You don't need to overhaul everything at once. Lucas: Exactly. And the data is already there. It's just a matter of using it.