Latest / The Retention Marketing Podcast with Fexingo: Email, SMS, and Loyalty for Repeat Customers / How a Wine Brand Uses Post-Purchase Survey Data to Trigger Repeat Emails
Transcript
- Lucas: There's a DTC wine brand based out of Sonoma — I'm not going to name them, because they're not public and they're pretty private about their data — but they shared some of their retention numbers at a conference last quarter, and one figure jumped out at me. Luna: I think I know which one you mean — was it the talk where they showed that email flow triggered by the post-purchase survey? Lucas: Exactly. So here's the setup: when a customer places their first order, they get an email the next day with a short survey. Five questions. Things like 'How did you hear about us?' — standard stuff — but then: 'On a scale of one to ten, how much did you enjoy the wine?', 'What's your preferred price point per bottle?', 'What occasion are you mostly buying for?' and 'How often do you drink wine?' Luna: So they're building a taste profile on day one. That's smart. Lucas: Right. And what they do with that data is the interesting part. They feed the answers into their email platform — they use Klaviyo — and they create segments based on the responses. So if someone says they buy wine for weekly dinner parties and they prefer bottles under $20, that customer goes into a flow that highlights their value-tier wines and sends a reorder reminder around day 21. Luna: That timing is key — the average wine drinker finishes a bottle in about a week, but they might buy in a three-pack, so 21 days is right when they'd be finishing their first order. Lucas: Exactly. And they track this. Their repeat purchase rate for customers who complete that survey is 45 percent within 60 days. For customers who skip the survey, it's 22 percent. So the survey itself is a retention lever. Luna: But that's not just the survey — it's the follow-up emails that are personalized based on the data. The survey is useless if you don't act on it. Lucas: Absolutely. And they act on it aggressively. They have six emails in that post-survey flow. The first one is a thank-you with a discount code for the next purchase — 15 percent off — which incentivizes completion in the first place. The second email comes three days later, and it's a product recommendation based on the taste profile. If they said they liked bold reds, they show a Cabernet or a Zinfandel. Luna: So they're not just guessing. They're using the data to make the next suggestion relevant. Lucas: Precisely. The third email is a reorder reminder timed to the drinking frequency they indicated. If they said 'weekly,' it's at day 21. If they said 'occasionally,' it's at day 45. The fourth email is a loyalty program invite — they have a VIP tier that gives early access to new releases. The fifth is a 'we miss you' with a stronger discount, and the sixth is a win-back with a last-chance offer. Luna: Six emails over two months — that's a lot. How do they avoid annoying people? Lucas: They limit the flow to customers who actually completed the survey, so there's an opt-in quality to it. And they suppress anyone who unsubscribes from the flow but keep them in the general list. Also, the emails are spaced out — at least three days between each. The conversion rate on each email drops off, but the cumulative effect is strong. Luna: I'm curious about the survey itself. If it's five questions, what's the completion rate? Lucas: They said around 35 percent of first-time buyers complete it. That's actually high for a post-purchase survey. I think the discount code helps — 15 percent off the next order is a decent incentive. Luna: Yeah, and the timing is smart. The day after delivery, the customer is excited about the wine. They're more likely to engage. Lucas: Right. And they also send a reminder SMS 48 hours later if they haven't completed it. That bumps the completion rate by about 12 percentage points. Luna: Speaking of things that make you stop and think — a lot of what we talk about on this show, from survey triggers to email flows, comes down to actually implementing the ideas. And if these conversations have sparked something you've tried or improved in your own business, that's exactly what keeps us going. This show is ad-free and listener-supported, so if you find value in these episodes, you can support us at buy me a coffee dot com slash fexingo. No pressure — just a way to keep the content coming. Lucas: Yeah, and we appreciate every bit of support. It lets us dig deeper into cases like this wine brand instead of chasing advertisers. Luna: Exactly. So back to the survey — one issue they ran into was data decay. People's tastes change. Someone who liked bold reds in January might switch to rosé in July. Lucas: That's a great point. They handle it by re-surveying customers every 90 days if they haven't made a purchase in that window. And they also update the profile based on actual purchase behavior. If someone buys a white wine after saying they only drink reds, the system overrides the survey data with the behavioral data. Luna: So the survey is the starting point, not the final word. Lucas: Right. And they found that customers who give a low enjoyment score — say, a 4 or below — are actually more likely to repurchase if they get a personalized apology and a recommendation for a different style. That's a retention win that's only possible because they asked. Luna: That's a counterintuitive insight. Most brands would just write off a low score as a lost customer. Lucas: Exactly. But by acting on the data, they can salvage that relationship. Their recovery rate for low-score customers is about 28 percent — meaning those customers make a second purchase after the apology email. Luna: That's significant. So the survey isn't just for segmentation — it's for service recovery. Lucas: Exactly. And it's automated. They don't need a human to read every response. The system flags any score below 5 and triggers a different flow: a sincere apology, a recommendation based on other survey answers, and a 20 percent discount code. Luna: So it's a win-win. The customer feels heard, and the brand gets another shot. Lucas: And the data feeds back into their product development too. They track which wines get low scores most often and use that to inform future sourcing. But that's a whole other episode. Luna: One thing I want to circle back to: the survey length. Five questions seems like a sweet spot. Do you know if they A/B tested the number of questions? Lucas: They did. They tested three, five, and eight questions. Three questions had a higher completion rate — about 45 percent — but the data wasn't rich enough to personalize effectively. Eight questions dropped the completion rate to 20 percent. Five was the balance point: 35 percent completion with enough data for six segments. Luna: So the six email flows correspond to six segments? Lucas: Not exactly. They have more segments than that based on combinations of answers. But the five questions give them about a dozen meaningful profile types. The six emails are the same for everyone, but the content changes based on the profile. Luna: Got it. So the structure is fixed, but the personalization is dynamic. Lucas: Exactly. And they track performance at the profile level. They found that the 'weekly dinner party' segment has the highest repeat rate — 52 percent — while the 'special occasion only' segment is at 18 percent. So they adjust the discount offers accordingly. Luna: That makes sense. The special-occasion buyers probably need a bigger incentive to convert. Lucas: Right. So they offer a 25 percent discount to that segment versus 15 percent for the high-intent segment. The ROI still works because the lifetime value is lower for that group, so they're not over-investing. Luna: One risk I see is over-segmentation. If you have too many segments, you might not have enough volume per segment to get statistically significant results. Lucas: That's a real concern. The brand said they try to keep each segment above 500 customers before they start personalizing offers. Below that, they use a general flow. So there's a threshold. Luna: Good discipline. So the takeaway from this episode is: a short, well-timed post-purchase survey, integrated with your email platform, can dramatically improve repeat purchase rates — if you actually act on the data. Lucas: And if you keep the survey fresh. Data decays, tastes change. Re-survey or update based on behavior. The brand we discussed does a quarterly refresh, and that keeps their segments accurate. Luna: It's a simple system that delivers outsized results. I think a lot of listeners could implement this within a week using tools they already have. Lucas: Agreed. The hard part isn't the tech — it's having the discipline to ask, and then to listen.