The Psychology of Referral Programs

Depending on where you look it up, the average Facebook user has somewhere between 100–200 connections. Social networks have enabled communication with acquaintances on a global scale, and the internet has also enabled online commerce. The combination of these two is where online referral programs live and become super interesting—and super measurable.

Every marketer’s dream is for their campaigns to go viral; no platform is cheaper and more effective than organic word of mouth. When it comes to marketing, customers recommending a product and converting their own friends to users is as good as it gets.

A referral campaign’s goal is to target an existing customer base—people who are happy with their experience—and inspire them to recommend the product to their friends. Referral programs accelerate organic word-of-mouth campaigns by offering incentives. The plan, in short, is to catch the wave and ride its momentum to success.

In 1999, author Dr. Ivan R. Misner talked about word of mouth as “the world’s most effective, yet least understood marketing strategy.” But a lot has happened in the last 20 years. Every click, every purchase, and every message sent over the internet can be captured and analyzed in order to assess a campaign’s efficiency.

What gets measured gets improved, so marketers can build ever-better programs. From PayPal to Dropbox to Uber, many companies rely on referrals as their main marketing channel. In the 20 years since Misner’s statement, surely we’ve all learned a trick or two.

What gets measured

Every company, product, and situation will be different, but in this age of programmatic advertising and online sales, marketers don’t have to guess. They can simply run experiments to test how their own customers will react to a campaign.

Advertisers run so-called A/B tests all the time to see which version of an ad is better. The campaign that generates more traffic or conversions will then be introduced to a bigger portion of the target group.

Using different adverts can have as much impact on the outcome as targeting different consumer segments with the exact same message. Big data helps finetune the message, the target group, and every little detail of the campaign. Since much of the data and research is freely available, we can use the information to develop a good starting point for our own experiments.

Framing the call to action

And here we have the marketer’s conceptual dilemma. Can they make a referral look like an organic, word-of-mouth type of invite?

Consider the following sentence: “You will both get 10 percent off your next order.”

1. By rephrasing it to, “You’ll get 10 percent off your next order. Your friend will also get 10 percent off their next order,” we haven’t changed the meaning of the sentence. The offer is still the same; the person sending the referral and their invitee each get exactly a 10 percent discount, yet in an A/B test, each message will usually yield different results.
2. Marketers can not only change the framing between egoistic and pro-social, but they can also change the offer itself. How would you feel about getting only 5 percent off if you invite a friend and they get a full 10 percent? How about the other way around? It turns out that the perceived “fairness” of the split can depend on how well you and your friend know each other.
3. Much of social media is monetized through a platform’s own advertising network, which offers detailed insight into a campaign’s performance. For acquaintances, a fair-and-square equal split can enhance a referral program’s performance, but it turns out that “fair and square” is not super important to close friends.
4. And to put another variable in play: Some companies pay customers for every new referral, and others apply a cap. Dropbox, for example, offers the same reward but has a cap of 28 referrals. Wise, on the other hand, employs a lower threshold: no payments for the first two referrals; only the third one comes with a reward.

A grain of salt

All of these strategies can work; the optimal incentive depends on the company, the product, and the audience—on the campaign in general. Every referral program needs to be customized based on industry characteristics, a product’s price, and a million other variables.

Marketers might find that their most successful framing depends on whether they target customers right after their purchase or in a later campaign. Some companies have found that financial incentives reduce the number of people purchasing the recommended product. And some campaigns work for repeat purchases, but not when the customer is interacting with the brand for the first time.

So here we go: The world’s least understood marketing strategy still confuses marketers today. But at least we have the tools to measure what we need to and steer the campaigns while they are still running. We don’t need to understand everything as long as it all just works in the end.