Not legal advice: This article covers marketing practice and general legal awareness of advertising-disclosure and consumer-protection rules around referral and affiliate rewards, not a legal audit. If you run a referral or affiliate program, confirm current disclosure requirements — especially outside the US — with a qualified lawyer.
"Your happiest customers are your best salespeople" is one of those lines nobody questions anymore, and it usually arrives with a specific set of numbers riding along: some fixed multiplier for how much more a referred customer is worth, a precise share of all purchases supposedly driven by word of mouth, a clean lift from switching to a two-sided reward. Some of that is real, matched, peer-reviewed research. A lot of it is a number repeated across referral-software vendor blogs that cite each other instead of a dated, disclosed study.
We went and checked: the Journal of Marketing study that tracked roughly 10,000 real customers of a German retail bank for almost three years to measure what a referral is actually worth, the 2018 follow-up that used a different sample from a similar bank to test whether the relationship between the referrer and the referred friend matters, the largest disclosed global survey on which advertising formats people actually trust, the peer-reviewed research on whether rewarding a referral helps or backfires, and the 2023 update to the US FTC's rules on disclosing a referral or affiliate reward — plus what Israeli consumer-protection guidance actually says about it. Several of the most commonly repeated referral-marketing numbers didn't survive the check; we flag those plainly instead of repeating them.
Key takeaways
- A Journal of Marketing study that matched roughly 10,000 real customers of a German retail bank by demographics and acquisition timing found referred customers' lifetime value was at least 16% higher than similar non-referred customers' — in both the short and long run, though the advantage was smaller for older or low-margin acquisitions.
- The relationship matters, not just the referral itself: a 2018 Journal of Marketing Research study of 1,800 referred customers and 3,663 other new customers at a German bank found referred customers were 40% more likely to leave once the person who referred them had already left — part of the retention gain is borrowed from the referrer's own loyalty, not a clean property of the new customer.
- Nielsen's disclosed, 30,000-respondent, 60-country 2015 survey found recommendations from people you know are the single most trusted advertising format (83%), ahead of branded websites (70%) and online consumer opinions (66%); a smaller, US-only 2021 follow-up (2,000 respondents) put the figure even higher, at 89%.
- A peer-reviewed 2007 Journal of Marketing study found that offering a reward does increase referral likelihood overall, but rewarding only the person making the referral — and not the friend receiving it — can reduce their willingness to refer, especially for well-known brands, because it raises doubts about the recommender's motive.
- The US FTC's June 2023 update to its Endorsement Guides treats an affiliate or referral link as a "per se" material connection that must be disclosed clearly and conspicuously, not buried in a bio. Israel has no referral-specific statute, but both the Attorney General and the Consumer Protection and Fair Trade Authority said in 2021 that compensated promotion must disclose the commercial relationship under the existing law against misleading consumers.
Are referred customers actually worth more?
The most rigorous answer we could find comes from Philipp Schmitt, Bernd Skiera and Christophe Van den Bulte, in "Referral Programs and Customer Value," published in the Journal of Marketing in January 2011 — a peer-reviewed journal article, not a vendor white paper. The data: roughly 10,000 real customers of a German retail bank, tracked for almost three years. To isolate the referral effect from who simply happens to arrive through a referral versus another channel, the authors matched referred customers to statistically similar non-referred customers — same demographics, same acquisition period — rather than comparing raw averages.
The result held up in both directions that matter: referred customers' lifetime value was at least 16% higher than their matched non-referred counterparts, in both the short run and the long run. But the size of that advantage wasn't uniform — it was smaller when the referral program was used to acquire older customers or lower-margin customers, which is why the authors' own recommendation is a selective referral program rather than a blanket one open to every segment.
16%+
the minimum lifetime-value advantage of a referred customer over a matched non-referred one, found in a peer-reviewed study of roughly 10,000 real customers of a German retail bank.
Source: Philipp Schmitt, Bernd Skiera & Christophe Van den Bulte, "Referral Programs and Customer Value," Journal of Marketing, Vol. 75 (January 2011), pp. 46–59
The 16%+ figure is a floor, not a single clean number — and it's made of two different trends moving at different speeds. The contribution-margin advantage of referred customers narrows over time, while the retention-rate advantage persists. That combination is exactly why the lifetime-value gap stays positive even years in: referred customers don't necessarily keep spending more per transaction, they mostly just keep being customers for longer.
Does the referral relationship itself affect retention?
A follow-up study by largely the same research group dug into a question the 2011 paper couldn't answer on its own: does it matter who referred whom? Christophe Van den Bulte, Emanuel Bayer, Bernd Skiera and Philipp Schmitt answered this in "How Customer Referral Programs Turn Social Capital into Economic Capital," published in the Journal of Marketing Research in 2018. The dataset: 1,800 new customers of a German bank who arrived through a referral, plus 3,663 new customers who became customers through other means, with the referrers' own records matched in wherever possible.
Two findings stood out. First, referrers and the friends they referred tended to have similar profitability — consistent with referrers acting as a matching mechanism, steering the bank toward people who were genuinely a good fit rather than referring at random. Second, and more specifically, customers referred by older, non-divorced, longer-tenured and already highly profitable existing customers went on to become more profitable new customers themselves than other referrals did.
| Study | What it actually found |
|---|---|
| 2011 — Journal of Marketing (~10,000 customers, tracked ~3 years) | Referred customers' lifetime value at least 16% higher than matched non-referred customers; the margin advantage narrows over time, the retention advantage persists; weaker effect for older or low-margin acquisitions |
| 2018 — Journal of Marketing Research (1,800 referred vs. 3,663 other new customers) | Referrer and referred friend tend to have similar profitability; customers referred by older, longer-tenured, highly profitable referrers became the most profitable new customers; referred customers were 40% more likely to leave once their own referrer had already left |
+40%
how much more likely a referred bank customer was to leave once the customer who referred them had already left — evidence that part of a referral's retention benefit is borrowed from the referrer's own relationship with the business.
Source: Christophe Van den Bulte, Emanuel Bayer, Bernd Skiera & Philipp Schmitt, "How Customer Referral Programs Turn Social Capital into Economic Capital," Journal of Marketing Research, Vol. 55 (2018), pp. 132–146; summarized in Harvard Business Review, September–October 2018
The authors theorize two mechanisms working together: better matching (a referrer already knows both the business and the friend, so tends to refer someone genuinely suited to it) and social enrichment (the social tie itself, not just the product, is part of what keeps the referred customer engaged). The 40% figure is the clearest evidence for the second mechanism — and the clearest warning that a referral program's retention numbers can look better on paper than they'll hold up once your most active referrers eventually churn themselves.
Do people actually trust a recommendation more than an ad?
Nielsen's Global Trust in Advertising survey is the largest disclosed, dated source we found on this: 30,000 online respondents across 60 countries, rating their trust in 19 different advertising formats. It's worth being precise about what this is — a large, disclosed, methodologically transparent commercial survey, not a peer-reviewed academic study, and one based on claimed trust rather than observed buying behavior.
The headline finding has held up across editions: recommendations from people respondents actually know are the single most trusted format by a wide margin, ahead of every paid, owned or earned alternative the survey tracked.
A smaller, US-only Nielsen follow-up in 2021 (2,000 respondents) put the "recommendations from people I know" figure even higher, at 89% — worth reading as a different-scoped data point (US only, a fifteenth of the sample size) rather than a clean trend line, but directionally consistent. As for the other commonly repeated line in this space — some version of "word of mouth drives 20–50% of all purchasing decisions" — it does trace to a real, named source: a 2010 McKinsey Quarterly article by Jacques Bughin, Jonathan Doogan and Ole Jørgen Vetvik, which also estimated that a high-impact recommendation is up to 50 times more likely to trigger a purchase than a low-impact one. Treat it as a real but dated estimate, not a current measurement — it's now over fifteen years old.
Does offering a reward actually increase referrals — or can it backfire?
Lalin Anik, Christopher Berry and others have studied referral incentives from several angles, but the paper most directly on point is Gangseog Ryu and Lawrence Feick's "A Penny for Your Thoughts: Referral Reward Programs and Referral Likelihood," published in the Journal of Marketing in 2007. Across controlled studies, the authors found that offering a reward does increase the likelihood someone will refer a friend — but the effect isn't uniform, and the direction can flip.
The reward helped most for weak ties (acquaintances rather than close friends) and for weaker, less-established brands. For strong ties and well-known brands, rewarding only the person making the referral — and not the friend receiving it — could actually reduce their willingness to refer, because it raises the "social cost" of the ask: the friend may start to wonder whether the recommendation is genuine or just a bid for a kickback. The authors' practical conclusion: for a strong brand, giving at least part of the reward to the friend, not just the referrer, tends to perform better than a referrer-only reward.
A cluster of specific multipliers circulates across referral-software vendor blogs on top of this — claims like "double-sided programs convert 2.4x better" or "referred customers are 5x more likely to buy." We looked for a disclosed sample size, date range or published methodology behind each one and came up empty; they trace back to a vendor's own aggregated customer data, presented without the kind of detail that would let an outside reader check the number. Treat the direction (reward both sides where you can) as supported by peer-reviewed research, and treat any specific multiplier attached to it as an unverified vendor claim.
Do you have to disclose a referral or affiliate reward?
In the US, the FTC's Endorsement Guides (16 CFR Part 255) were updated in June 2023 — the first substantive update since 2009 — specifically to address modern referral links, affiliate codes and influencer-style promotion. The rule: when a connection between the person making the recommendation and the business "might materially affect the weight or credibility" of that recommendation, and the audience wouldn't reasonably expect it, it has to be disclosed clearly and conspicuously. An affiliate or referral commission counts as a material connection by definition, regardless of the amount — and the guidance explicitly says a business can no longer rely on a bio-only disclosure or a buried "click for more" link.
The FTC also has a specific enforcement mechanism here worth understanding precisely: in October 2021, it sent "Notices of Penalty Offenses" to more than 700 companies, putting them on formal notice that fabricated or undisclosed endorsements are an unfair or deceptive practice. That notice itself doesn't mean a recipient did anything wrong — but a company that was notified and then violates the rule can face civil penalties, which started at $43,792 per violation in 2021 and, adjusted annually for inflation, now run close to $53,000. This is a different legal mechanism from a straightforward trade regulation rule: it depends on a company having been formally put on notice first, not an automatic penalty that applies to every business from day one.
Israel has no statute written specifically for referral or affiliate disclosure. What exists instead: in January 2021, the Attorney General's office took the position, in a court proceeding, that undisclosed sponsored or influencer content on social media can mislead consumers under the general deception prohibition in the Consumer Protection Law, 5741-1981. In July 2021, the Consumer Protection and Fair Trade Authority followed with its own public clarification that paid or incentivized promotion — which a referral or affiliate reward clearly is — must disclose the commercial relationship, explicitly describing this as a clarification of existing law rather than a new policy or new standard.
In practice, the disclosure duty falls on two places at once: the person making the public recommendation (a WhatsApp status update or a social post asking friends to use their referral link, not a private, one-to-one message) and the business running the program, which should give referrers simple, ready-to-use disclosure language rather than leaving them to guess. Israel's guidance is non-binding interpretation rather than a standalone statute, but it points at the same underlying behavior the FTC's rule targets — so treating referral promotion as something that needs disclosure, in either market, is the safer default.
How do you build a referral flow that pays out on real purchases, not just signups?
- 1Give each customer something specific and easy to share — their own tracked link or a short personal code — rather than a vague ask to "spread the word." The research above is about a concrete referral being traceable back to a specific person, not generic advocacy.
- 2Capture the referral the moment a new contact actually arrives: a Webhook trigger from your referral landing page or checkout, or a Field changed trigger the moment a "referred by" field gets set on the new contact's record.
- 3Don't release the reward on signup alone. Wait for the Sale recorded trigger — which fires the moment a purchase is logged to your Sales log — on the referred contact, then use a Condition node to check that their "referred by" field is populated before paying out. Rewarding a completed purchase rather than a click is what both bank studies above actually measured, and it keeps the incentive aligned with revenue rather than noise.
- 4Reward both sides where you can, not just the referrer — the 2007 study above found a referrer-only reward can quietly reduce willingness to refer, particularly for an already well-known brand.
- 5If cost needs a ceiling, use a Capacity gate (e.g., a bonus pool for the first 100 referrals this month) rather than silently lowering the reward value later once a program catches on.
- 6Send the reward as an email with automatic WhatsApp fallback if it goes unopened, and hand referrers a one-line disclosure to use if they post publicly about the program — the simplest way to stay inside both the FTC's and Israel's current guidance.
None of this needs a dedicated "referral" feature in oToK — it's the same coupon-code mechanics covered in our coupon-code-marketing article for the reward itself, chained to the Sale recorded trigger and a Condition node so the payout only fires once the referred friend has actually become a paying customer, not merely clicked a link.
Frequently asked questions
- Are referred customers really worth more, or is that marketing folklore?
- It's real and peer-reviewed: a Journal of Marketing study that matched roughly 10,000 customers of a German retail bank by demographics and acquisition timing found referred customers' lifetime value was at least 16% higher than similar non-referred customers', in both the short and long run — though the advantage was smaller for older or lower-margin acquisitions, which is why the authors recommend a selective referral program rather than a blanket one.
- Does a referral program really improve retention, or is that overstated?
- On average, yes — with an important asterisk. A 2018 Journal of Marketing Research study of 1,800 referred bank customers found they were 40% more likely to leave once the customer who referred them had already left, meaning part of the retention benefit is borrowed from the referrer's own relationship with the business rather than being a clean, standalone property of the new customer.
- Should I reward the person referring, the friend they refer, or both?
- The peer-reviewed evidence favors rewarding both where you can. A 2007 Journal of Marketing study found offering a reward does increase referral likelihood, but rewarding only the referrer — not the friend — can reduce their willingness to refer, especially for a well-known brand, because it raises doubts about the recommender's motive. We couldn't independently verify the specific multipliers ("2.4x better," "5x more likely to buy") that circulate across referral-software vendor blogs; treat those as directional vendor claims, not peer-reviewed findings.
- Do I legally have to disclose a referral or affiliate reward?
- In the US, yes: the FTC's Endorsement Guides, updated in June 2023, treat an affiliate or referral commission as a "per se" material connection that must be disclosed clearly and conspicuously, not buried in a bio — and the FTC can pursue civil penalties (now close to $53,000 per violation) against companies that were formally put on notice in 2021 and then violate again. Israel has no referral-specific statute, but the Attorney General and the Consumer Protection and Fair Trade Authority both said in 2021 that paid or incentivized promotion must disclose the commercial relationship under the existing law against misleading consumers.
- What's the best channel to pay out a referral reward?
- We didn't find a large, disclosed, referral-specific study comparing delivery channels the way review-request research exists for reviews — it hasn't been independently measured at this scale. What the research above does support is paying out after a real purchase (the Sale recorded trigger, not a signup) through a reliable path — an email with automatic WhatsApp fallback if it goes unopened keeps the reward from quietly getting lost, which matters more here than any one channel's edge.
