Latest / The Retention Marketing Podcast with Fexingo: Email, SMS, and Loyalty for Repeat Customers / How a Coffee Brand Uses SMS to Boost Repeat Rate by 40 Percent
Transcript
- Lucas: So there's this coffee roaster — we'll call it Morning Fuel, because they asked me not to use their real name — and they had a pretty classic problem. Luna: Let me guess: a lot of one-time buyers, not enough repeat orders. Lucas: Exactly. They had a solid email program, but the repeat rate was stuck around 18 percent for customers who bought a bag of whole-bean coffee. And coffee is a replenishment category — you'd think people would reorder naturally. Luna: But they didn't. So what did they try? Lucas: They decided to test SMS. Not as a replacement for email, but as a separate channel for a specific subset of customers: anyone who opted in during checkout. And they built a three-message sequence that's worth walking through. Luna: I'm all ears. What was message one? Lucas: Message one went out 24 hours after delivery. It was a thank-you, but not the generic kind. It included a brewing tip tailored to the roast they bought — like, 'If you're using a French press, try a 4-minute steep for this Ethiopian Yirgacheffe.' Personal, useful, no ask. Luna: That's smart. It's a value-add, not a sales pitch. And it reinforces that they know what they're doing. Lucas: Right. Message two was timed based on bag size. They knew a 12-ounce bag lasts about 14 days for a typical two-cup-a-day drinker. So on day 14, they sent a text: 'Your bag is probably getting low. Want to reorder? Here's a link.' No discount, just a convenience play. Luna: And that worked? Lucas: It did for about 12 percent of the SMS group. But the real magic was message three. If the customer hadn't reordered by day 30, they got a text that said: 'It's been a while. We miss you. Here's a free shipping code on your next bag.' And they signed it with the roaster's name, not the brand. Luna: Wait — 'signed with the roaster's name'? Like a person? Lucas: Yeah. The actual person who roasted that batch. They tested two versions: one automated from 'Morning Fuel' and one signed '— Jason, your roaster.' The human-signed version converted 15 percent better. Luna: That's a huge lift from just a signature. I wonder if it's because we're so used to automated texts that a real name cuts through. Lucas: That's exactly what they found in their post-purchase survey. Customers said the human touch made it feel less like a marketing blast and more like a personal check-in. Now, the overall result: within 90 days of implementing the sequence, their repeat purchase rate among SMS opt-ins jumped from 18 percent to 25 percent. That's a 40 percent relative lift. Luna: Forty percent. That's hard to ignore. But how do they know it was the SMS that caused it, not just a coincidental increase in loyalty? Lucas: They ran a proper incrementality test. They randomly split their SMS opt-in list into two groups: a test group that got the sequence, and a holdout group that got no SMS at all — they still got the regular emails. The test group's repeat rate was 25 percent; the holdout was 18 percent. So the lift was real. Luna: Good. That's the kind of rigor more brands should use. What about the risk of annoying people? Coffee drinkers don't want texts at 7 AM. Lucas: They were careful about that. They sent messages between 10 AM and 2 PM, and they gave every customer a preference center link in the first text. You could choose frequency: 'weekly tips,' 'only reorder reminders,' or 'special offers only.' Only about 3 percent unsubscribed after the first message. Luna: So the opt-in was strong. And the preference center probably made people feel in control. That's a subtle but important trust signal. Lucas: Exactly. And they didn't stop at three messages. They've since added a fourth for customers who haven't ordered in 60 days — a 'we've got a new roast' alert with a link to a tasting note video. But that's still in testing. Luna: I love that they're iterating. But let me ask you this: why do you think SMS outperformed email for this audience? Morning Fuel's customers skew younger, right? Lucas: Yeah, their core demographic is 25 to 40, urban, mobile-first. Email open rates for that group were around 22 percent; SMS open rates were 78 percent. And the time to click was faster — most clicks happened within 30 minutes of sending. For email, it was more like 4 hours. Luna: That speed matters for a reorder reminder. If you're out of coffee and thinking about it, a text that arrives right then can capture the impulse. Lucas: Right. And they found that the reorder reminder text — message two — had the highest conversion rate of the three. About 8 percent of recipients clicked and bought. That's pretty strong for a no-discount message. Luna: So the key takeaway might be: give people a useful reason to reorder, not just a discount. The brewing tip built goodwill, the timing reminder solved a problem, and the human-signed we miss you added a personal touch. Lucas: That's the framework, yeah. And it's replicable for any replenishment product — not just coffee. Pet food, skincare, vitamins, even razor blades. The principles are the same: know your consumption cycle, add value before you ask for the sale, and let a real person sign off. Luna: I think that last point is underrated. We've all gotten those automated texts that feel like a robot trying to be friendly. A real name changes the feeling. Lucas: It does. And it costs nothing to implement. You just need a CRM that lets you pull the name of the person who handled that order. Most platforms can do that. Luna: Alright, so if someone listening wants to try this, what's the first step? Should they just start texting their whole list? Lucas: Absolutely not. First, make sure you have a clean opt-in process. Morning Fuel used a checkbox at checkout with a clear explanation: 'Get brewing tips and reorder reminders via text.' No pre-checked boxes. Then start with a small test segment — maybe 10 percent of your opt-ins — and run a holdout group to measure incrementality. Without that, you're flying blind. Luna: And set up the preference center from day one. That's probably the biggest lesson here: give people control over frequency, and they'll stay engaged longer. Lucas: Yeah. Oh, and one more thing — they used a short code, not a long number. That helped with deliverability and trust. Customers are more likely to respond to a 5-digit code than a 10-digit number they don't recognize. Luna: Good point. Short codes cost a bit more, but for a program like this, it's worth it. Lucas: So to wrap up the case: Morning Fuel saw a 40 percent lift in repeat purchases from a simple three-text sequence, driven by personalization, timing, and a human touch. And they proved it was incremental. That's a solid win for a channel that a lot of brands still underutilize. Luna: I'd add that it's also a win for the customer. They got useful information when they needed it, not spam. That's the kind of retention marketing that builds loyalty, not just transactions. Lucas: Absolutely. And you know, if this episode sparked an idea you actually used — maybe you set up a preference center or tested a human-signed text — that's the kind of thing that keeps this show ad-free. If it was worth a coffee to you, we're at buy me a coffee dot com slash fexingo. No pressure, just a way to keep these conversations going. Luna: Yeah, we hear from listeners who say they've applied something they learned here, and that's the best feedback we could ask for. Every bit of support goes straight into making the show better. Lucas: Alright, back to the coffee. One thing I didn't mention: Morning Fuel also tested a 'subscribe and save' offer in their SMS, but they found it cannibalized one-time purchases without increasing overall revenue. So they dropped it. That's a reminder that not every tactic works — you have to test. Luna: Interesting. So even within a successful program, you still have to be willing to kill what doesn't work. Lucas: Exactly. And that's the mindset we'll keep exploring. Thanks for listening, and if you have a retention story you want us to dig into, let us know.