# Refer-a-Friend Offer on the Thank You Page A/B Test

Page: https://alidemirbas.com.tr/en/lab/ab-test-library/tests/refer-a-friend-offer-on-the-thank-you-page-ab-test

## Hypothesis

If the thank-you page shows one defined refer-a-friend offer, some buyers who have just finished an order may share or copy their referral link. That could raise the referral sharing action rate, though a card that pushes ahead of the order summary may make the confirmation details harder to find.

## When to run this test

Give this test a slot when buyers show signs of recommending you after a purchase, such as survey answers saying they would recommend the product, product links being shared on their own, or support requests asking about a referral program. The pattern to look for: buyers explicitly share or copy the link after seeing the offer, and people arriving through it place new orders within the follow-up window. Before launch, make sure you can actually deliver the promised reward, state its terms clearly and describe its value without exaggeration.

## Primary metric

**Referral sharing action rate**: Measure an explicit share or link-copy action per eligible assigned buyer; opening a share menu does not prove the invitation was sent.

## Other metrics

- **Purchase rate from referral links**: Measure completed orders from eligible visitors arriving through shared referral links over the same follow-up window.
- **Referral link use**: Track link-copy or share actions among assigned buyers; actual delivery may not be observable.
- **Reward redemption rate**: Compare rewards used among eligible recipients over the same follow-up period.
- **Order confirmation clarity**: If surveyed, use the same short question and sampling method in both arms to check that order details remain clear.

## What to check before and during the test

- **Offer**: Add one predefined referral invitation and keep any incentive terms fixed.
- **Sharing controls**: Keep the available channels and their order the same.
- **Timing**: Use one fixed post-purchase display trigger.
- **Presentation**: Use one dismissible card format.
- **Device**: Verify the selected sharing behavior on each device before launch.

## What to watch when reading the result

- Opening the share menu does not show the link was sent; count only explicit share and link-copy actions. If sharing goes up while purchase rate from referral links and reward redemption rate stay flat, the offer's real effect may be limited.
- Check that buyers who dismiss the referral card, or never see it, can still reach the order summary and tracking details. If order confirmation clarity drops, do not count the rise in sharing as a win.
- Verify that the referral link is never shared automatically on any channel without the user's consent; that would also inflate the result. The finding applies to the offer, reward terms and channel order you tested; confirm the sharing behavior works the same on each device and read device results separately.
- If the gap in referral sharing action rate is small, enter each arm's assigned buyers and explicit share actions into the [A/B test significance calculator](https://alidemirbas.com.tr/en/calculators/ab-test) to judge whether it could be sampling noise. Orders from referral links are rarer and arrive later; until the follow-up window closes, do not report an unclear result on that secondary metric as no effect.

## Frequently Asked Questions

### Does opening the share menu count as sharing?

No. It does not prove delivery of the link. Measure explicit share or copy-link actions per eligible assigned buyer and disclose any limits on observing actual transmission.

### What should be tracked after the share action?

Track completed new orders from eligible referral-link visitors over a common window, plus reward use among entitled customers. More sharing without new orders may have limited value.

### How can the invitation avoid competing with order details?

Keep order summaries and tracking accessible. Verify access in usability checks and monitor relevant events and support contacts. Direct comprehension claims require appropriate feedback, but every test does not need a survey.

### What should the referral offer make clear?

Offer a reward you can fulfill, explain its conditions and avoid exaggeration. Keep incentive type and sharing-channel order fixed. Never share automatically without the user's permission.