How to Blow Up Repeat Revenue

Some products create repeat customers. Others create one-time buyers. Most marketers can't tell the difference because they measure success in 7-day attribution windows.

A couple of years ago, I created the following analysis for an ecommerce business in order to boost repeat purchase rates. What started as a simple investigation into product performance became the foundation for a tactic that can easily deliver 30-50% increase in CLV if executed correctly. And it literally takes half an hour to implement.

The analysis revealed something counterintuitive: their bestselling products weren't creating their best customers.

Here is the ranking I generated (details anonymized and different products used for illustration):

The numbers tell an interesting story. The Hoka One One Rincon 2, at the top, shows an 8.26x multiplier—meaning customers who bought this shoe first generated over eight times the baseline average order value in their first 12 months. The Brooks Transcend 7 follows at 8.06x, while the Adidas Yeezy Boost 350 comes in at 8.01x.

Meanwhile, at the bottom of the list, products like Dunk Low Crazy Camo show just 2.7x multipliers. These were some of their most heavily promoted items based on conversion rates and margins, but they were creating one-and-done customers.

The difference between an 8.26x and a 2.7x multiplier means you can afford to pay 3 times more to acquire a Hoka Rincon 2 customer than a Dunk Low customer and still come out ahead on CLV.

The pattern revealed that certain products served as relationship builders, while others were merely transaction generators. The Hoka One One Rincon 2 was not their bestseller and did not yield their highest margin. However, customers who initially purchased that specific running shoe developed a relationship with the brand that was worth eight times their initial purchase.

What made this discovery powerful wasn't just identifying these patterns—it was understanding that we could systematically calculate these multipliers using statistical models.

This playbook will show you exactly how to create your own Product Hitlane ranking using the same statistical methodology and—more importantly—how to use it to transform an acquisition strategy from gambling on conversions to investing in repeat purchases.

A Marketer's Optimization Trap

Most ecommerce marketers optimize acquisition campaigns for the wrong metrics. They chase immediate conversion rates, promote bestselling products, or focus on the highest-margin items for top-of-funnel advertising. This approach leaves massive CLV potential on the table.

Traditional acquisition thinking:

• "Our bestseller converts at 8% - let's promote that"
• "This product has 60% margins - perfect for ads"
• "Our hero product gets the most clicks."

The hidden cost of this approach is acquiring customers with poor long-term retention characteristics while missing products that create loyal, high-value customers.

Consider two acquisition scenarios:

Scenario A: Traditional Bestseller

• Conversion rate: 8%
• AOV: $75
• Customer repeat rate: 25%
• Lifetime value: $120

Scenario B: Hero Retention Product

• Conversion rate: 6%
• AOV: $65
• Customer repeat rate: 47%
• Lifetime value: $185

Even with lower immediate performance, Scenario B delivers 54% more lifetime value per customer. At scale, this compounds dramatically.

The Solution: Hero Retention Products

Hero Retention Products are items that, when purchased first, create customers with significantly higher repeat purchase rates and lifetime value than your baseline expectations. These aren't necessarily your bestsellers, highest-margin products, or most popular items. They're the products that create the most valuable customer relationships.

While competitors can replicate an ad creative, copy product descriptions, or match prices, they cannot copy your understanding of which products create valuable customer relationships without doing the same deep analytical work. This knowledge becomes your sustainable competitive moat.

The Methodology

The analysis requires a single transactions CSV file with the customer's purchase history.