Episode Transcript
Will: Hello and welcome to the Advocacy Channel, a customer marketing podcast. I'm your host, Will Fraser, a customer marketing expert, entrepreneur, and the co founder and CEO of sasquatch. This is our very first episode and today we're chatting with Rachel Gershon, the assistant professor of marketing at the Ratey School of Management, University of California, San Diego, to chat about her research on recipient versus sender benefiting rewards programs. Hi Rachel. Thank you very much for joining us today.
Rachel: Thanks so much for having me.
Will: This is our first podcast episode. We're really excited to dig into it. We've got some great conversations to have here. But just to help give some context to our listeners, can you tell us a little bit more about your role as an assistant professor and what does that look like?
Rachel: Sure. I teach a course called Consumer Behavior to undergrads at UC San Diego. I like to say that this course is an intro site course with commercials. But in addition to teaching a course, most of my time is spent doing research and I get to kind of study whatever interests me within the realm of human decision making. So I do work on donation behavior and voting. How do you motivate people to exercise? And then I also have a lot of work on referral rewards.
Will: Cool. And how did you get into that? How did you find that space? That's not necessarily something that everybody just naturally thinks, you know, it's a five year old, I want to go and research that. Like, how did you get there?
Rachel: So few people aware that you can do a PhD in marketing. I certainly didn't know that you could. So I was an undergrad and I was interested in psychology and nonprofit development. And I found a woman who later became my advisor, Cindy Kreider. And I became her research assistant. I was very interested in her work and financial decision making. And I ended up just sticking around and doing a PhD with her. And I then eventually became a professor after my five year PhD program. But again, very few people know that this is a career path and it's a really cool one. So I highly recommend people looking into it who are interested in research.
Will: I enjoy the casual PhD in that story there where just casually, just not that it's a major accomplishment or anything, I just thought I would get that.
Rachel: Oh man. Every year at the PhD, I was like, I'll just do a couple years and then I'll find my real job and stuck around.
Will: Yeah. And that is amazing. Like you said, you know, we work with marketers all the time and it's not very common to kind of find that PhD in marketing. But more importantly, the insights that you're able to kind of get from this research, I think is really the kind of gold that so many of the marketers we talk to are looking for. Right. It's very - we want to be data driven. We want to be able to know what the best practices are. And all too often the best practices are closer to common practices and the data is closer to theories. And so we really are challenged quite often to get concrete data on this. So I'm really excited to kind of dig into, to understand a little bit about some of your findings.
Rachel: Well, thank you. Yeah. And we end up pulling a lot of our research from places like sasquatch and just other findings from companies that are willing to publish it. So there's definitely a lot of backend and forth between academia and practice.
Will: Yeah. And that's how, I mean, in my world, that's how it should be. I think that we can maybe sometimes get a little bit overly confident that we actually know how things work. Right. Just because we've done them a certain way is not really the same as knowing. And I think that academic partnership with the insights from the commercial world leads to some really good findings.
Rachel: Absolutely.
Will: And on that, you know, I'd love to dig into this paper that you did around pro social referral incentives. So maybe you could just kind of give us a high level overview of kind of what you guys were trying to figure out and what was the hypothesis that you were exploring.
Rachel: So we were interested in these pro social referral incentives, which we can also refer to as recipient benefiting incentives. So this is the idea that you have a current customer and rather than giving them a reward, you say, hey, you can give your friend a reward to join our service or buy our product.
And we were interested in whether there were contexts in which this kind of referral incentive would be effective. Now, very few companies use this kind of referral incentive, so we didn't have a lot of faith that this would be a broadly effective referral scheme. And initially we were specifically interested in whether this would work for sensitive contexts. So, you know, medical context. You know, we were working with a specifically reproductive health clinic.
Will: Okay.
Rachel: And we were wondering whether, you know, given that they get a lot of their patients through word of mouth, they could offer their current patients the ability to give their friends a reward for joining the clinic. And we thought maybe this would be a particularly important incentive scheme for sensitive context because people don't want to gain something if there's potentially like important consequences.
We ended up not working with this reproductive health clinic for various reasons. They weren't able to run the study. And we ended up working in kind of broader contexts and found that these pro social incentives were broadly effective. And they turned out to be just as effective as a shared incentive in a lot of contexts.
Will: So maybe we can just step back there kind of. If you say what is, you kind of mentioned that quickly what a pro social incentive is, you know, it's an incentive that's going to really help the person you're referring. The alternative is then a different type of incentive.
Rachel: Yeah. So we call them pro social and selfish incentives. And then we're kind of adding like a motivation in that case to the incentive. And that's why we changed it to recipient benefiting and refer benefiting throughout the paper. Because it doesn't necessarily have to be - you don't have to have a selfish motivation in order to refer your friend and receive a reward. But that's maybe like an inference that people could make if, you know, if I referred Will to join a service and I get money, it's possible people will make selfish inferences. So there's a selfish incentive in which the current customer receives reward, a pro social incentive where the recipient receives reward, and then the most common is a shared incentive in which both the referrer and the recipient receive rewards.
Will: You're kinder than I, I call it a greedy reward. So selfish is even a little bit nicer than that there.
Rachel: Totally. Yeah.
Will: It's a really interesting question you're asking, right? Is, you know, is this pro social going to deliver better results than selfish intrinsically? You know, I think we would assume, you know, just as average humans, that selfish is probably going to deliver more than pro social. But it sounds like you found at least that they're the same, if not the opposite.
Rachel: Yes. What was interesting is a lot of, you know, when people think about creating these incentive schemes, you think about getting the ball rolling by getting your customers to refer. Right. So in order to get someone to do that first action, we use incentives. So it makes a ton of sense that we want to create an incentive to get that first step of the referral process going. How do you get your customers to spread word of mouth? But of course there's two steps to the process, right? There's getting your customer to refer and then also getting the recipient to follow through on the referral and what we were able to find with our studies is that actually offering an incentive can have just as much - provide just as much motivation for making a referral as offering an incentive for their friend. And that's because there's two different factors involved. Either your new customer, your current customer gets a reward for themselves or they receive a reputational benefit. So either you get a reward or your friend thinks that you're a nice person. And we actually don't know through our studies whether your friend really thinks that. But we know that refers think that their friends will think more highly of them. So we have these two factors that might be contributing to the referral behavior. And then we have the recipient behavior and the recipients have the harder job, right? They either have to purchase a product or you know, sign up for a mailing list and you know, you already get a million emails every day. And because the recipients have that harder job, the incentive is more important on their side. So what we find is there's no difference at the referral stage. And then recipients are more likely to follow through if they receive a reward.
Will: You see, I like the way you're breaking it down, right? Because this is something that no one's ever broken it down like this for me. And I mean I've been working in referral programs for 10 years. And just to break it into that separate, to say we're going to have to start them, we're going to get them to make a referral, but then we actually have this action stage where someone has to do something. And quite often in my experience, I've either seen we call them double sided, right? But we see both the sender and the recipient receiving a reward or we see just the sender. This is a really, really interesting idea to me to say that it's actually better to - if you were only going to reward one side, you might want to think about rewarding just the recipient and not the sender at all.
Rachel: Yeah, I think one of the reasons we thought the exact same way when we were starting this process with the reproductive health clinic, when we started this, we thought that what was important was the referrals motivation, the referral's motivation. We really weren't thinking about that recipient. And one of the reasons I think that we all tend to focus on that first step is that it's really hard to get data on both steps separately. And what we tend to find when working with companies is just the conversion rate. So you give people one incentive or another and you see how many people follow through on the referral because you're giving people a code and they can refer through emails or text messages or in person. And all you can see is that final number of the conversion rate. So we got really lucky that in one of our studies, the company that we were working with was actually sending an email through their like through their new customers or through their original customers. So they'd say, give us your friend's email address and we'll send the email.
Will: Right.
Rachel: And therefore they were able to see how many referrals did they send and how many people followed through. And we had all three steps of the process. But it's really hard to get that those middle pieces and that's the only way we were able to see how this is breaking down.
Will: Yeah, it's interesting. Yeah. Because we see in our data we can see the start, we can see that they press the button. But due to a lot of GDPR and data considerations, we often go black, black a little bit in the middle and then we come out the other side and we can see you visited a web page, you signed up, you converted. But on that note of conversion, did you see any difference in conversion rates? Is there a plus or minus to that pro social versus selfish reward around an ultimate conversion rate?
Rachel: Yes. So the ultimate conversion rate was higher for the prosocial reward. And we had two field experiments and a few hypothetical lab experiments and actually one incentivized lab experiment as well. We used those to be more controlled. You want to be able to randomize, you know, who are you sending this referral to? So it's not that people are sending referrals to different friends if they receive a selfish or pro social reward. So we do actually find that the conversion rate is higher for a pro social referral. We didn't find a difference between pro social and these double sided or shared referrals. And there's actually currently working paper, so it's not yet published, but there's a paper that's out there that I've seen presented where they found something similar, but they actually doubled the size of the shared reward. So in that case I can either send my friend $10 or we each get $10 or I get $10. So in that case, one of them is double the cost of the company and they still found no difference between the pro social and the shared. So it could be that you could make a smaller incentive for a pro social and that people are somewhat insensitive to incentive size when they're passing their friend a reward.
Will: Interesting.
Rachel: Yeah, really interesting finding.
Will: That's, that has some big ramifications like literally half the cost of their, you know, their referral program. And we already see referral programs even with double sided often being 40% or less below like their cost per say like a paid acquisition. So much cheaper already. But that could be a pretty game changing finding as well now I guess.
Rachel: Interesting. It tends to be about 40% of the cost of a paid marketing version.
Will: Yeah. So if you were buying like Google Ads, let's say so we will see that a strong program is going to be about 40% cheaper than that channel. Right.
Rachel: Is a great statistic that I'm keeping.
Will: Yeah, yeah. And that's, I mean I will say that's what we found. We are not professors here. We are practitioners of the field. But site.
Rachel: Well, yeah, yeah.
Will: But that's really interesting to see that the conversion rate doesn't really shift too much when you're changing that reward. And does the propensity to actually make a referral, it sounds like it's relatively unaffected or does it even increase when you have a pro social reward? How does that look?
Rachel: So at least what we found across our studies is that there was no difference between I receive a reward or my friend receives an equal size reward and actually no difference between for me getting half that reward too. So I give you 10, I received 10 or I received five. We did not find any differences. And our explanation is that you assume that there's some reputational benefit to passing your friend a reward going into it. We thought that there might be more of a discomfort for sending a referral where your friend didn't receive a reward. But we didn't find that that played as big of a role in this context that it really was. I want my friend to think highly of me and that this could help.
Will: You know it's funny because there's data from, I want to say it's from the 1980s, some data kind of indicating that actually paying cash versus other rewards can actually disincentivize a friend to friend referral because they start to recognize that they've become a salesperson is the hypothesis of course is what they kind of theorize from that. And I think we've seen that internally. We have seen obviously some catches to this. And it's kind of where my brain's going is we will see cash based incentives really effective inside of financial products where people are think - our working theory is they're thinking about their money. I'm in here about saving money, investing money, whatever. Building my credit and therefore money is really my main thought. And we will find that money as a - I mean, it's like a cash deposit, typically not. So cash payout can be a strong incentive there, but in other places we've seen it not work as well. So I guess with this, my question is, you know, is there, are there any situations where you think that this pro social, selfish reward dynamic doesn't hold? Are there any types of referrals or products that you think it may not hold in?
Rachel: That's a good question. We haven't tried like products as opposed to cash. All of ours were cash or gift cards. And I do share that theory that there could be contexts in which cash could make people less comfortable where you don't want to be a salesperson to your friends. And I think that that's a good theory. In terms of times when the pro social reward might backfire a little bit, I could imagine that there is a boundary at which point, like for the size of the incentive, where you're like, okay, why does my friend get $500 for joining this apartment building or becoming an employee at this company and I receive nothing? I think that there could certainly be a point in terms of incentive size. That's not something that we found. We did sort of a reasonable product service incentive sizes, maybe 5 to $30. But I could certainly imagine a size at which people expect to also receive some reward. Other possible boundary conditions, given that we find most of the effect on the recipient side, one way that we were able to turn the effect off is by decreasing the action costs on the recipient side. So just making it super easy to join or really low cost, you know, we're really not expecting almost anything of you as a new customer might decrease the need for that pro social incentive. Yeah.
Will: And are there any cases - I'm just kind of thinking of this as a marketer trying to use this where I'm like, you know, did we find that this works better in certain communities or when I'm referring certain people, or is there any way that pro social is good here, but really you're not going to get the same benefit in another way? Or was it all pretty much across the board? It worked.
Rachel: Yeah. We really were looking. We looked at looks. We looked for like age differences, gender differences, really context differences too. You know, like what if the - it's a more sensitive context or something along those lines. We didn't find a difference interesting in these contexts. I'm excited for this other paper to come out. We'll certainly send it to you once it's published.
Will: Please do.
Rachel: Yeah, yeah. But I think that they tested several other contacts and so hopefully they're coming out with some new boundary conditions and other future work can look into that. But we did not find contexts where pro social incentives backfired.
Will: Cool. Now there's a question that we get asked about every once in a while, which is where people want to do a charity donation in somebody's name. So they want to donate - you know, here you refer a friend and when they sign up, we'll donate $5 in your name or their name. Did you find that that kind of an incentive was useful? You know, we have very limited data on it, so I'm interested in your insights for sure.
Rachel: I'd love to hear what you guys have found too. We, I have tried so hard to find a way to make donation incentives effective and all I have in this published paper is that, you know, fairly low power study, which is our study1 in this paper where we did a donation incentive and we found that it wasn't any more effective than the control where there was no incentive. In other studies that are unpublished, we have tried donation incentives in a lot of different contexts and we just don't find that they, that they're effective. Unfortunately, I continue to have hope that there are contexts where it might work and I know of incentive schemes where people do a double sided referral reward and donation incentive and maybe that makes everyone feel even better than just a donate double sided reward. That's not something I've tested, but I hold out hope that there are ways to make that work and we haven't found it yet in our studies.
Will: Yeah, our experience is somewhat limited, but one of the things we worked with one of the large online shoe retailers, okay. And they believed that donations could be very effective because there's a well known online shoe retailer that does kind of a buy a pair, give a pair kind of program. And they thought that that might be really interesting and it was not highly effective. That and our working theory, and this is a absolute working theory I'm sharing because that's a smart thing to do, share publicly is that basically if the brand is aligned with a charity, right. If this is already a reason that you connect and purchase with this brand, then double down on the charity component in the referral program because you know, you know whether this is a brand where when I buy one of their shirts, they plant tent trees. Or if this is, you know, a charity that wants to donate a pair of shoes for every pair I buy, that's maybe part of why I'm part of it. And maybe this can help me more versus I buy shoes because I like shoes and - oh, I see. We could donate to the SPCA or whatever it might be and yeah, that's cool and all, but I kind of like shoes. So totally working theory, no academic backing on that. But we have seen, in practice, we haven't seen highly effective results either. So always curious on that.
Rachel: You know, all we're looking at is how effective is this in terms of referrals? I don't know how effective it might be in terms of brand benefits. It might be that that makes your new customer, your customers feel really good about shopping with you and doesn't increase referrals because, you know, they want a reward, they want to pass a reward to their friend, but it might make them happy in other ways with the brand. Yeah, and that makes that, that also makes me think, you know, in terms of other things that we can look at with referrals, we like to look at does this increase referrals and conversions, but we've only recently started looking at does this increase valuable new customers.
Will: Right.
Rachel: And that's been an exciting piece of just having more data and getting to look at customer activity in addition to whether they were referred. And I think that that's going to be like the next big thing for which referral reward schemes are effective.
Will: And I think that what you're touching on is a question that's really not been asked or answered that much yet, which is really, what are the ancillary benefits of referral programs? Because everyone is so focused, whether we're working with business to businesses or direct to consumer businesses, they're very focused on show me the provable revenue from this program. And obviously that's an important part of it. You should do that. But, you know, we see things like if you get someone to make a referral on Twitter, okay, let's say you will drive substantially more visitors to your website than if they were to share that via Facebook messenger or via, you know, other channels. Now, if your goal is conversion, I can tell you that that traffic that's generated from some of those other channels like Twitter, on average, converts less. However, we've worked with some clients where their goal is to grow their retargeting audience. Some goal, some is it's brand exposure. So for some people, they're like, you know what, it's okay. We're a slightly more complex purchase. We know they need to see us five or six times. And this is just one step in that journey. You know, there's questions that I saw very many questions around that. And then like you said, there's this question of what does it do to lifetime value? You know, what does it actually do to that ongoing value there? And we've seen some companies we work with, Bank, Digital bank, where they realized that their program was fairly easy for someone to get the reward. And so people were very rapidly getting the reward before they had actually seen and felt the value of the service. And so then they would really have no reason to stay as a customer. So that conversion point was too early.
Rachel: Basically in the journey, we found something similar with a large field partner where we were excited to see, you know, most referrals happen within the first few weeks of joining the, of joining the service. But those are not the most valuable referrals. So in terms of when do we offer an incentive? You know, they have this valuable incentive program and people tend to make their referrals early on. They might not start offering an incentive until you've been with the company for a month because those early - and those early referrals tend to be people who want the reward and aren't necessarily very familiar with the product, don't know which friends to refer. So there's other things to think about besides the conversion, but those things tend to be a little bit harder to calculate.
Will: Absolutely. And it's interesting. I mean, I'm excited obviously about this entire field, but to me, we see quite often people asking the question of should I just offer this referral program to everyone, you know, anyone who comes to the website or whatever can make a referral, or should I only offer this to active customers or a specific group of partners? And it sounds like with this case at least as one example you're talking about, really the answer is not even just all customers. It might be actually to customers that have hit a certain mark where you're now comfortable that they're going to properly bring in valuable new customers to your business.
Rachel: Yeah, it's an interesting thought to like segment your customer - like, perhaps they'll offer the ability to refer customers using a pro social reward or not offering any incentive at all when, when you originally joined and then, you know, once you're a loyal customer. Certainly there are some company companies like Blue Apron doesn't offer the ability to refer unless you've, you know, used the product for a while and then they were actually always our example as the pro social reward because there aren't that many pro social referral rewards. But Blue Apron offers the ability to send your friends meals for free once you've been with them for a certain period of time.
Will: Yeah. And I think that that's, that's a really important piece and it's a piece that we're trying to work on on our side right now, which is how do we develop advocates? Right. So a lot of this starts with the idea that, well, we have happy customers and they're excited and amazing and let's get them to make referrals. But we're kind of currently, our kind of internal hypothesis is that actually you are developing advocates through that journey. Right. And you may use incentives as part of that journey to actually develop them. So that they're kind of ultimate state is this idea of being an advocate and then, you know, how do we keep them in that state? So I love where this is going. Some really interesting point. Yeah.
Rachel: Then, you know, there is a paper that finds that just the act of referring makes people more committed to the company. There's maybe like some cognitive dissidents there where you're like, okay, if I told my friend to join this company, I must-
Will: Well, yeah, yeah, no, no, it absolutely makes sense. You know, we always think about it. People don't want to be, people don't want to be perceived as liars. And if I told you it's good, canceling is the opposite of the truth there. So.
Rachel: Well, and that's actually, I don't have a ton of evidence for this, but we did find some evidence in our lab experiment that people who received a selfish reward actually said they liked the product better. And remember, these are just like randomized into conditions. They all actually took a personality quiz and were asked if they wanted to send it to their friend and if they received the reward. Rather than giving it to their friend or splitting it, they said that they enjoyed the personality quiz more. Which is kind of a funny little bit of psychology. Yeah, you're like, okay, I must love it if I'm, if I'm receiving a reward. I wouldn't send them something with no reward at all if I didn't enjoy it.
Will: That's interesting. So if you can basically, you know, I'm paying maybe more, maybe double sided here in this case, I'm selfish and pro social and I'm giving the pro social because I know it will increase the take up rate from the actual recipient. But I'm actually giving the current customer reward because it's a loyalty incentive and not a referral incentive.
Rachel: Possibly.
Will: Yeah.
Rachel: And I'd love, you know, we should look at more, you know, field data for this before feeling very confident about it.
Will: Yes, yes.
Rachel: This was like a weird lab personality quiz where I think that there's more ability to move people on how much they enjoyed it than maybe a product that they've, that they've really experienced. But I think it's possible that there are other benefits to providing a reward for your customers. What we did look at with the self benefiting versus other benefiting rewards and that's I guess the fourth term I've used.
Will: Yeah.
Rachel: So I apologize.
Rachel: But what - one thing we were able to look at was whether people stayed with a company after being offered these rewards. And we didn't find a difference there. So at least that's one metric of whether, you know, being offered these rewards changes your perception of the company or how much you like the company. But looking at other metrics of do you continue to be an active user? Would be, would be a helpful next step.
Will: And that was the recipient, the new customer didn't stay on as a customer for less or more time when being referred. They kind of had an average lifetimes that you're saying there.
Rachel: This was looking at the referral. Okay, sorry for the confusion. This was looking at whether the refer was more likely to stay with the company. Just to see, you know, okay, we're offering these two different referral rewards. They don't seem to have a different effect on their likelihood of referring. Does it have any effect on how much people like the company? Doesn't seem to, at least with this one metric. But there are certainly other interesting metrics that we could look at.
Will: Yeah, no, that's, this is, this is the stuff that fascinates me. And I mean, just on a quick side note, I think this is talking about referral programs. If you want to confuse a English native speaker, attempt to explain the difference, you know, a referrer, referee, recipient, new customer. It is a space just ripe with confusion in the terms. But yeah, this is all really, really fascinating stuff and we're talking a lot about what might be interesting to go and study. Are you working on anything right now? Is there a big project underway that you can give us a little bit of a sneak peek at?
Rachel: Sure, yeah. I'm working with a colleague at Wharton on a project right now where we just find with a couple field partners that people who join through a referral are also more likely to refer. And I briefly mentioned this to you, and you said that you've also seen this. It seems to be something that is maybe clear to people who are in this practice of creating referral programs. But we're interested in not only are people more likely to refer, but why. And sort of the uninteresting explanation would be that people who join through a referral are part of, like a social network that is more likely to refer. That's more extroverted. But a more interesting explanation for us would be that joining through a referral somehow makes it more socially acceptable to send a referral in this same context. And what we find in more controlled lab experiments is that when you randomly assign people to say, like, you joined this company through either a referral or through an ad, they become more likely to send a referral if they joined through a referral.
Will: Really?
Rachel: Yeah. And we look at why, and there's a few reasons, and it seems to be, you know, they feel that it's more appropriate. They feel lower psychological costs, so less discomfort, and they feel that the product will be a better fit. So there's a little bit of like a social value of having joined through a referral originally. And but we also see in the data that this decays over time. So once you've been with a company for four or five months, having joined through a referral does not have as large of an increase in your likelihood of referring.
Will: Interesting. So this is an interesting one for me because I've seen a lot of referral programs where they launch with a really high incentive. Right. And then they slowly bring the incentive value down.
Rachel: Right.
Will: And I'd be curious if you couldn't, you know, do the math to show that actually when you launch a referral program, we want to go for really, you know, whatever is the highest converting it may not make the most financial sense because we're going to start that. We're going to like, we're kind of starting the flywheel. We're pushing the boulder uphill. And once we get it started, then we're going to be able to drive this, what we call the culture of referrals around getting those people. I'm - Obviously, we'd have to look into it.
Rachel: But I like the culture referrals. So that's interesting. So you're saying when a company, like, with a startup, they might start with the largest referral incentive just to get the most possible customers who join through referrals yeah, that does seem to be what this data would show is just having people who join through referrals makes it more likely that, that they will refer. They feel more comfortable doing it. And it's in that specific context which I find so interesting. Right. Like people are offered referrals through friends all the time, but - but that doesn't seem, it seems to be somewhat important that they join through referral in this context for this company.
Will: Yeah, it's interesting. We've - we've often taken the hypothesis, which is not a very interesting hypothesis, but that simply people who have been referred know there is a referral program, which is one of the great challenges that faces so many companies where their referral program is internally. They don't necessarily see it as a benefit to the customer. They see it as maybe this thing that marketing has pushed in or this idea that we have to do, but people really are just here for the product, which we do not see to be the case. We see that when you make a referral program highly visible, people participate. That's always been our idea. But I really like your possible explanations as well there that the idea that it is just more socially acceptable because, well, when they've done it themselves, they maybe they feel that's more socially acceptable. Yeah, that's interesting.
Rachel: That's actually a great idea for - an alternative explanation would be that, yeah, they just know about the referral program, which is why it's nice to have these kind of contrived but very controlled lab experiments where we just say like, okay, imagine that you joined through a referral image imagined you joined through an ad. You now have an opportunity to refer your friend. Would you like to do it? And in that case, they both know about the referral program. Another piece of evidence that I think helps with that alternative explanation is that we do control for the amount of activity that both referred customers and non referred customers have with our big data sets. Because one other explanation would be that people who are referred just are more valuable users, they're more active. And that's also true. There's previous research that finds that our data also shows that people who join through a referral tend to spend more, tend to have more activity. But even controlling for all of that, all of that use of the product, that might lead them to see that there's a referral reward program. Even controlling for that, we still find an increase and referrals.
Will: That's awesome. That's awesome. It's wonderful to be able to talk about this. You know, a lot of this is data, some of this is hypothesis still. If you were kind of, you know, from this, this research and your experience, you know, if we were going to kind of hand over a few kind of key takeaways to, so a product manager or marketer who's like been tasked with launching the referral program and they're trying to figure out this, this incentive structure problem which everybody is trying to get right. Kind of what are the big takeaways you would offer to them?
Rachel: Yeah, I think that one thing I can say with some amount of confidence is offering the incentive to the recipient. So whether that's a double sided or a pro social referral reward, there seems to be a real benefit pretty consistently to offering an incentive to the recipient. And my biggest piece of advice was to the extent that you're, is, to the extent that you're able to just experiment, you know, if you can take 5, 10% of your customer base and ab test different referral incentive schemes, it's, it's a lot easier to know or you can say with a lot more confidence about your customer base, what do they respond to? Then you know, we can say, we have, we have great evidence for several different contexts. But it's nice to know within your customer base before launching the entire incentive scheme what's going to be the most effective. So experiments, Love experiments.
Will: Yeah. And it's interesting, I love this data that we're finding that you're bringing here about these pro social at least double sided programs because I will tell you, we speak to companies all over the world all the time where there's real concern from the management operations side about doing double sided.
Rachel: Yeah.
Will: And what we normally hear is someone say, well we're going to start with just the recipient or sorry, just the sender. And what I'm hearing here is that we should be fighting tooth and nail to say, hey, if you want to start with only one side, start with the recipient. And that's, that's a really interesting tell there, a really interesting piece of data there. Well, Rachel, this has been amazing. This is wonderful.
Rachel: I can't.
Will: It's been wonderful. I can't wait to have you back on to talk about the next paper that comes out, the next piece of data and absolutely have loved this. So thank you very much for your time.
Rachel: Thank you so much for having me. This is great.
