Latest / The Email Marketing Podcast with Fexingo: Newsletters, Open Rates, and Subscriber Growth / Why Email Segmentation by Behavior Beats Demographics
Transcript
- Lucas: You know, we talk a lot about list hygiene and deliverability on this show, but I think one of the biggest missed opportunities I see is how people segment their lists. Luna: Oh, absolutely. And I bet most people default to the same few categories: age, gender, maybe location. Lucas: Exactly. Demographics are easy because they're often collected at sign-up. But the data that actually predicts what someone will click on is behavioral — what they've bought, what they've browsed, how they engage. Luna: So you're saying I shouldn't be sending different emails to men vs. women? Lucas: Not entirely — it can be a starting point. But I want to dig into a specific case. There's a mid-size e-commerce brand I follow — they sell home goods, think kitchen tools and decor. They used to segment by age and gender, pretty standard. Then they switched to behavioral triggers: last purchase date, product category viewed, email engagement history. Luna: And what happened? Lucas: Over three months, their click rates went up 34 percent. Revenue per email rose 22 percent. And their unsubscribe rate actually dropped, because people weren't getting offers for things they'd never buy. Luna: That's a pretty dramatic shift. But isn't behavioral segmentation harder to set up? You need decent data and probably some automation. Lucas: It is more work initially, but it doesn't require enterprise tools. Even basic email platforms like Mailchimp or Constant Contact let you create segments based on purchase history or link clicks. You don't need machine learning. Luna: Right, so the barrier is more about knowing what to look for. What's the simplest behavioral segment someone should start with? Lucas: I'd say 'last purchase date.' Anyone who hasn't bought in 90 days is a different customer than someone who bought last week. You can send re-engagement offers to the first group, and cross-sells or thank-you content to the second. Luna: That's almost too simple. But I bet most people aren't doing it. Lucas: They're not. And the same goes for 'pages visited.' If someone browsed your site and looked at cookware but bought nothing, you can send a follow-up focused on cookware, not your entire catalog. Luna: That feels more relevant, and less spammy. I've personally unsubscribed from brands that send me everything because I bought one thing once. Lucas: Exactly. And that's the core problem with demographic-only segmentation: it assumes people in the same age group want the same things, which we know isn't true. A 35-year-old who buys baking supplies is different from a 35-year-old who buys power tools. Luna: So the behavior reveals the intent, not the demographic. Lucas: Right. And the nice thing is you can layer them. Start with behavior, then use demographics as a secondary filter. For example, send a cookware campaign to everyone who browsed cookware, then A/B test subject lines by age group. That's fine. Luna: Let's talk about engagement as a behavioral signal. I know open rate isn't perfect, but if someone hasn't opened an email in six months, they're probably not going to. Lucas: That's a great point. And it ties back to our episode on sunset policies. You should have a segment for 'inactive subscribers' and treat them differently — maybe a last-chance email, then suppress them. It protects your sender reputation. Luna: It also saves money if you're paying per send. Lucas: Exactly. So behavioral segmentation isn't just about increasing revenue; it's about reducing waste. And I think that's the argument that convinces leadership. Luna: We should probably mention that this doesn't mean you need a perfect data warehouse from day one. Even simple rules help. Lucas: Totally. Start with one behavioral segment. Test it against your usual demographic blast. See if the numbers move. They almost always do. Luna: I'm going to try the 'last purchase date' one this week. Honestly, if today's conversation sparked something you'll actually use in your own email marketing, that's great. And if you find it valuable enough that a coffee feels fair, there's a link: buy me a coffee dot com slash fexingo. It helps keep the show ad-free and focused on stuff like this. Lucas: Yeah, we appreciate that. And it's a small way to support the work we put into each episode. No pressure at all. Luna: Anyway, back to segmentation. One thing I've seen work really well is combining behavioral and demographic when you have enough data. Lucas: Absolutely. For example, a clothing retailer could segment by 'browsed men's shoes' and then further by 'age 25-34' if they have different style preferences. But the behavior is the primary driver. Luna: And you can automate this with most email platforms now. There's really no excuse not to try. Lucas: Yeah, the technology is there. The hurdle is mindset. Many marketers still think in terms of 'who' instead of 'what they do.' Luna: So if someone is listening and wants to start tomorrow, what's one actionable step? Lucas: Go into your email platform and create a segment of subscribers who have purchased in the last 30 days. Then create another of those who haven't purchased in 90 days. Send a different email to each group. Measure the click-through rates. I'd bet money the behavioral group performs better. Luna: And if you don't have purchase data, use click data. People who clicked on a specific link in the past are telling you their interests. Lucas: Perfect example. So to wrap up: stop guessing who your subscribers are based on their age or gender. Watch what they do. That's the real signal. Luna: And if you try it, let us know how it goes. We'd love to hear if the numbers shift for you. Lucas: Until next time.