The Situation
A Swiss online retailer had an email list of 4,200 subscribers collected over three years — and was doing almost nothing with it. The only emails going out were monthly newsletters with a discount code, sent manually, with no segmentation. Open rate was 14%, click-through rate was 1.8%.
The store had no automated sequences. A customer who signed up received nothing until the next newsletter. A customer who abandoned a cart heard nothing. A customer who purchased once was treated identically to someone who had never bought anything.
Revenue from email: roughly 4% of total store revenue. Industry average for a well-run e-commerce email programme is 25–35%.
What I Did
Audit and platform assessment. I started with a full audit of the existing list — subscriber age, engagement history, purchase behaviour, and bounce rates. 31% of the list hadn’t opened a single email in 18 months. I ran a re-engagement campaign to that segment before building anything new; unresponsive contacts were removed. A clean list of 2,900 engaged subscribers is worth more than a bloated list of 4,200 mixed ones.
Welcome sequence. I built a five-email welcome sequence triggered the moment someone subscribes. The sequence introduces the brand over eight days — not with discounts, but with the story behind the products, what makes them different, and social proof from existing customers. The fifth email offers a first-purchase incentive. Conversion rate from subscribe to first purchase: 11%.
Abandoned cart recovery. Three-email sequence triggered one hour, 24 hours, and 72 hours after cart abandonment. The first email is a simple reminder with no discount. The second addresses common objections (shipping time, return policy). The third offers a small incentive. Recovery rate: 18% of abandoned carts resulted in a completed purchase.
Post-purchase flow. A four-email sequence starting immediately after purchase — order confirmation, shipping update, delivery confirmation with care instructions, and a review request 14 days after delivery. The review request email has a 34% open rate and generates the majority of the store’s product reviews.
Win-back sequence. Customers who hadn’t purchased in 90 days entered a three-email reactivation sequence. The sequence acknowledges the gap, highlights new products, and offers a loyalty discount on the final email. Reactivation rate: 9% of lapsed customers made a purchase within 30 days.
Segmentation. I restructured the list into five segments based on purchase history and engagement level. Newsletter content is now tailored per segment — new subscribers see different content than repeat buyers, and high-value customers receive early access to new products.
Results — 90 Days After Launch
- Email revenue share: 4% → 28% of total store revenue
- List open rate: 14% → 31%
- Click-through rate: 1.8% → 6.4%
- Abandoned cart recovery rate: 0% → 18%
- Welcome sequence purchase conversion: 11% of new subscribers
- Monthly revenue from email (90-day average): +340%
- Unsubscribe rate: below 0.2% across all sequences
Email became the store’s highest-ROI channel — higher than paid social, higher than Google Shopping, and with zero cost per send beyond the platform subscription.
What Made the Difference
The biggest shift was moving from broadcast to behaviour-triggered email. A monthly newsletter goes to everyone regardless of where they are in their relationship with the store. Automated sequences respond to what a customer actually did — signed up, abandoned a cart, made a purchase, stopped engaging.
Behaviour-triggered emails consistently outperform broadcasts because they arrive at the right moment, with relevant content. An abandoned cart email sent one hour after abandonment catches the customer while the purchase intent is still active. A welcome email sent the moment someone subscribes catches them at peak interest.
The list cleaning was also essential. Sending to unengaged contacts damages sender reputation with email providers, which means even your best emails land in spam folders. A smaller, engaged list outperforms a large, stale one on every metric that matters.