3 Proven Case Studies in Referral Program Design: The Complete Guide to 'Referral Marketing' for D2C E-commerce to Escape Soaring CPAs

For D2C e-commerce businesses struggling with referral program design. In an era of soaring CPAs, we explain the complete procedure for referral marketing to turn existing customers into advocates, using proven case studies from Dropbox, Uber, and Oisix, along with the AARRR model and a 100-day fan conversion checklist for funnel designers.

The cost of acquiring new customers is skyrocketing once again.

E-commerce businesses that have relied solely on advertising must have seen their profits steadily eroded over the past few years.
—This is why referral marketing (referral programs) is being re-evaluated in 2026.

In this article, from the perspective of a funnel designer, I will systematically explain 'referral program design' to turn existing customers into advocates and simultaneously increase new acquisitions and LTV.

The structural checkmate of the era of soaring CPAs

In auction-based performance advertising, the unit price increases as the number of bidders grows.

Moreover, targeting accuracy itself is declining due to cookie regulations.

In other words, the current reality is that CPAs are 'easy to rise and hard to make effective.'

The framework to keep in mind here is the AARRR model.

Referral is the fourth stage of AARRR.

McClure's argument is that acquisition via referrals is the 'foundation for monetization' that should be addressed after setting up Acquisition and Retention.

Fighting only with advertising means fighting only in the first stage of AARRR.

With this approach, profits will shrink every time the CPA rises.

Why are customers acquired through referrals completely different from those acquired through advertising?

Customers acquired through referrals are not just 'cheaper in acquisition cost'.

Retention rates, average order value, and loyalty all tend to exceed those acquired through advertising.

The same trend has been confirmed in B2B sector research.

Why is it so effective?

The reason is that the act of "referring" carries the referrer's own reputational risk.

If a friend is disappointed by a referred product, the referrer's personal relationship is damaged.

Those who are willing to take that risk to refer someone have a deeper level of passion than mere repeat customers.

This is the boundary between [Good Customer -> Fan] as described by funnel designers.

The moment they make a referral, that customer is promoted from a "buyer" to a "co-conspirator."

Winning patterns of referral programs seen in 3 proven examples

Referral programs that function on a global scale share a common design philosophy.

Example 1: Dropbox (Cloud Storage)

The key is that the reward is not "cash" but the "value of the service itself."

The more capacity they have, the harder it becomes to let go of Dropbox, simultaneously driving both referrals and churn prevention.

Example 2: Uber (Ride-sharing)

With an acquisition cost of $40 and monthly usage of $95, the payback period is within one month.

If you apply this to e-commerce, it means that if you design a referral program where the CPO can be recovered in one month, you can break free from dependence on advertising.

Case Study 3: Oisix (Food Subscription)

Because it circulates among customers with high homogeneity, such as homemakers, the design makes it easy to achieve a high CVR via referrals.

The key points are: "no limit on the number of referrals" and "the guest's benefit is more generous than the referrer's."

The design philosophy of "not that you benefit, but that someone important to you benefits" is excellent.

Checklist for Incorporating a Referral Program into the 100-Day Fan Conversion Plan

When I enter a site as a funnel designer, I decide to "definitely introduce the referral program at least once by the 100th day after the initial purchase."

The reason is to request a referral at the moment when customer satisfaction is at its highest.

▍Check 1: Expose the referral program in package inserts, LINE, and emails at three points: 30, 60, and 100 days after purchase.

▍Check 2: Provide both "cash discounts" and "service value" as incentives, allowing customers to choose based on their preferences.

▍Check 3: Make the benefit for the referred person (guest) more generous than the benefit for the referrer.

▍Check 4: Make the referral link sendable within 3 taps.

▍Check 5: Always provide feedback to the referrer on the "results of their referral."

Almost all reasons why referral programs fail to grow are either "not being promoted" or "the process is too cumbersome." It is a design problem, not a product problem.

Summary

In the era of soaring CPAs for D2C e-commerce, referral marketing is no longer a "nice-to-have" measure, but a "prerequisite for design."
Turning existing customers into referrers means promoting them from "buyers" to "accomplices."
Only businesses that can create that step will be able to break away from advertising dependence and transform into profitable e-commerce operations.

If you are going to start tomorrow, first decide to "expose the referral program at the three points of 30, 60, and 100 days after purchase."

That alone will bring the outline of your 100-day fan conversion plan into focus.
─Related articles: Please also see "What is F2 Conversion Rate | 3 Proven Examples to Boost LTV in Single-Item Repeat E-commerce" and "What is a Fan Community | The Mechanism that Increased LTV by 253% in D2C Cosmetics."