The idea behind both
People tend to trust recommendations from people they know more than they trust advertising. Referral and affiliate programs try to encourage those recommendations by rewarding them. The two are related but differ in who is doing the recommending.
- A referral program usually rewards existing customers for introducing friends and family.
- An affiliate program usually works with outside partners, such as bloggers, review sites or creators, who promote your products to their own audience and earn a reward for resulting sales.
In practice the line blurs, and many stores use one system for both. What matters is understanding what you are paying for and how you will know it worked.
Common reward structures
Two-sided rewards. Both the person referring and the new customer get something, such as a discount or credit. This feels fair and gives the newcomer a reason to try you.
One-sided rewards. Only the referrer is rewarded, perhaps with store credit.
Percentage commission. The partner earns a share of each qualifying sale, commonly used with affiliates. The percentage is your choice and depends on your margins.
Flat fee per sale or sign-up. A fixed amount for each new customer, which is simpler to budget.
Tiers. Rewards increase with the number of successful referrals. This can motivate active partners but makes the program harder to explain and manage.
Non-cash rewards. Free products, early access or loyalty points. These may cost you less in cash, though they still cost something.
Working out what you can afford
Rewards are a form of marketing cost, so judge them like one. Start with the profit you make on an order, then decide how much of that you can give up while still coming out ahead.
Here is a hypothetical example. Suppose a first order averages 40, your costs for the goods, packaging and payment fees come to 24, which leaves 16. If you offer a reward of 5 for a successful referral, you keep 11 on that first order. If the new customer also buys again, the later orders may be worth more. If they do not, you have still not lost money on the first one. A reward of 20 on the same order would leave you at a loss of 4 on the first purchase, which only makes sense if you are confident that enough customers come back, and that is never certain. These numbers are illustrative only.
This is where understanding your profit margin, average order value and customer lifetime value pays off. A more generous reward is easier to justify when customers tend to return, and harder when most buy once.
Setting rules that are clear and fair
Spell out in writing:
- what counts as a qualifying referral (for example, a first-time customer with a completed order);
- when the reward is earned and when it is paid, bearing in mind the return window. Many stores wait until a returns period has passed so that rewards are not paid on orders that are sent back;
- how the reward is delivered, such as credit, a discount code or a payout;
- limits, such as a maximum number of rewards per person;
- what happens if an order is refunded or canceled;
- grounds for removing someone from the program.
Keep the terms simple enough that a customer can understand them in a minute.
Disclosure and honesty
When someone is rewarded for a recommendation, the people reading or hearing it generally deserve to know. Many places have advertising rules that expect clear disclosure of a material connection between a promoter and a brand. The specifics vary by place, so check the rules that apply to you and to your partners, and ask a qualified adviser if you are unsure. This is general information, not legal advice.
Some practical points:
- Ask partners to disclose clearly. A short, plain statement near the recommendation, rather than something buried in small print, is the general idea.
- Do not script false claims. Partners should describe your products truthfully, and you should not encourage statements you could not support yourself.
- Be careful with reviews. Offering rewards in exchange for positive reviews is risky, and many places or platforms restrict it. Customer reviews are more useful and more trustworthy when they are genuine and include the critical ones.
- Tell customers how referral links work if you track them with codes or cookies, and describe this in your privacy policy.
- Do not use pressure. Messages that nudge customers to spam their contacts can harm your reputation. Sending messages to people who have not agreed to hear from you may also break anti-spam rules in some places.
Tracking and fraud
Every program needs a way to attribute sales, usually through unique links or codes. Check that your system records them correctly, including when a shopper clicks a link on one device and buys on another.
Watch for abuse. Common problems include people referring themselves with fake accounts, posting codes on coupon sites that were meant for friends, or placing orders only to collect rewards. Reasonable safeguards include limits per person, checks for identical addresses or payment details, and the delay mentioned above before rewards are paid. If a pattern looks suspicious, order screening can help you review it before you pay.
What to measure
- number of referred customers and orders;
- profit per referred order after the reward;
- whether referred customers return more or less than others;
- return and refund rates for referred orders;
- cost compared with other marketing, such as paid advertising.
Compare against what you would have spent elsewhere to win a customer. If a referral program brings customers at a cost you would accept from any other source, it is working. If it mostly rewards people who would have bought anyway, consider adjusting the structure.
Starting small
A modest first version, such as a single clear reward for referrals through a simple link or code, is enough to learn from. Affiliate partnerships often need more attention: finding partners whose audience fits your products, agreeing terms in writing, and checking their content. Outcomes depend heavily on your product and audience, and you should not count on any particular result.