A/B testing is one of the most powerful tools for improving the performance of your referral programme. This article explains how strategic A/B testing can significantly increase new customer acquisition – up to 4x in the first six months.
You’ll learn how to use the Benchmarking feature in the Mention Me platform to compare referral metrics with industry peers, identify areas for improvement, and prioritise tests.
You’ll also discover A/B testing ideas tailored to each part of the referral journey — from increasing impressions and enrolments to boosting shares, conversions, and purchases.
Impressions per order
The ratio of offer impressions to the number of orders. For every order tracked by Mention Me, how often is referral shown?
- This metric should be above 100%, indicating referral visibility outside the post-purchase overlay.
Recommendation: Promote the referral programme more widely.
- Do you have a homepage promotion live? This can drive up to 60% of referral acquisitions.
- Download a report of customers who enrolled but haven’t shared, and remind them to share.
Enrolment rate
The percentage of customers who enrolled in the referral programme compared to total impressions.
- Customers can only enrol once per offer. Multiple impressions do not equal multiple enrolments.
Recommendation: Promote referral beyond post-purchase overlays.
- Send regular email promotions to keep it top-of-mind for customers.
Share rate
The percentage of eligible customers who share the referral offer.
Recommendation: Focus on referrer incentives and sharing methods.
- Test increasing the incentive, lowering minimum spend or adding a bonus (e.g. a chance to win). Request a test setup.
- Reorder or trial different sharing options (e.g., native share vs. name share).
Shares per sharer
A measure of how frequently the average sharer shares with others.
Recommendation: Encourage sharing with more than one friend.
- Test copy that uses plural language: “Share with friends”.
- Launch competitions e.g. “Get a chance to win a gift card” for those who share with multiple friends. See ideas here.
Responses per share
Compares how many share responses (e.g., link clicks, name searches) are generated per share.
Recommendation: Encourage more referee interaction.
- Allow customers to personalise the message they send when sharing the offer to increase engagement.
Incented per response
Out of the customers who responded to a share, how many entered their email and received an incentive?
Recommendations:
- Add social proof (e.g. Trustpilot scores/reviews) to increase trust and motivation.
- Test CTA variations: e.g., “Get 10% off” vs. “Get 10% off our award-winning furniture”.
- Add USPs on registration page like: “100-day free returns”, “5★ Trustpilot rating”, awards won.
Purchases per incented
How many incentivised referees go on to purchase?
Recommendation:
- Reinforce your brand’s value on the fulfil/promise page: highlight service guarantees or other benefits.
Purchase rate
Percentage of shares that result in qualified new customer purchases.
Recommendation: Optimise the referee experience.
- Test shorter incentive validity periods: e.g., 14 days vs. 7 days to drive urgency.
Running A/B tests is vitally important to improve the effectiveness of your scheme.
Video introducing Benchmarking Intelligence
Video showing Benchmarking Intelligence feature