Make It Sell Itself: On Referral Systems

Reading Time: 12 minutes

One year into running FeedbackPanda, we released a referral system. It was an immediate success, and it stayed that way ever since. When we sold the business, around 40% of new signups came through our referral system.

Immediately after turning on the user-facing parts of the referral system, we started seeing results. Our social media channels quickly filled up with our customers crafting posts that contained their referral links. Signups went up, and churn rates went down.

In this article, I will explain what referral systems are, how they benefit everyone involved, and how you can set one up efficiently.

---

The FeedbackPanda Referral System

FeedbackPanda had been humming along with a steady influx of new customers for over a year. Through the strong word-of-mouth marketing that happened organically within the teacher tribe, our customers were used to their peers talking about our service.

We decided to add a referral system when we raised our prices by 50% on January 1st, 2019. The referral system was supposed to both amplify existing word-of-mouth marketing as well as giving our customers a means to offset the increased price of their subscriptions. It was, in essence, a retention play. That’s why we went with a double-sided reward structure: both the referring and the referred customer would get a substantial reward for participating.

It worked like this: for every three customers you referred, you would get a free month of the service. It would be automatically credited to your account the moment the third referred customer purchased a subscription. All referred customers would get a 50% lifetime discount on their subscription. After increasing our prices, this brought their monthly cost right back to where it was before but made the price so much more attractive compared to a non-referred subscription.

This massive discount had an interesting effect: prospects who had heard of our product actively asked for referral codes on social media and were often bombarded with large numbers of responses, each containing a referral link.

As a consequence, we saw our referral links pop up in lots of places: on Facebook, YouTube, Twitter, even in obscure subreddits and on niche blogs. It was a big success, and signups have been substantially referral-driven ever since.

So how did we get there? Let’s look into how referral systems work, what benefits good ones provide to everyone involved, the risks to your business and your customers, and finally, how you can set up a referral system and get it running.

How Does a Referral System Work?

Referrals are not a cheap marketing trick. When done right, they can be a great way to grow your business and provide value to new and old customers alike.

The idea is simple: you get a customer to invite a new prospect to your product, and you reward either just the referring customer or both for participating in the transaction.

Most often, a unique referral link is generated for each of your customers so you can track who invited whom. You then encourage your customers to share this link wherever they think new prospects might find it.

The reward of a referral system will vary significantly depending on the kind of service you offer. Dropbox famously used a referral system that gave both the advocate and the referred user increased storage space. At FeedbackPanda, we discounted the subscriptions of both parties. There are many more incentives you can use, such as store credit, coupons, cash, or gift cards.

The math is simple, too: if the discounted advocate revenue plus the discounted revenue from the referred customers is higher than the non-discounted revenue for a user that doesn’t participate in the referral system, your system works.

How Can a Referral System Provide Benefits

Referral systems are economically sensible because the rewards you provide are worth the increased revenue that is generated by customers who invest their time and energy into marketing your product.

This is particularly true in highly tribal niches, which is why we could offer a 50% discount and still significantly impact our MRR with the FeedbackPanda referral system.

When you sell into an interconnected group of people, the built-in virality will give a lot of visibility to community members who recommend your product. By offering a referral system, you add a way for new customers to get substantial savings on a recommendation by a trusted peer. You can’t do much better when it comes to incentivizing a conversion.

It becomes a win-win-win situation. The advocate gets rewarded for referring the new customers: both from the service and in the form of trust inside their tribe. The referred customer receives a reward for trusting the advocate. And your business gets a new customer that is inclined to trust the service. This kind of trust projection strengthens business relationships. In the end, this will positively affect your customer retention.

What Benefits Does a Referral System Provide to Advocates?

There are three kinds of benefits that stand out for the customers that refer your service to their peers.

The most visible one is the reward, often some form of cost reduction or improved access to the service. That is something the advocate may enjoy immediately after referring a new customer. This immediacy is essential: it allows customers to turn referring new customers into a planned and scheduled activity.

Our customers at FeedbackPanda were Online English Teachers. As it turned out, the Chinese online schools they were teaching for had understood the value of a referral system too. The schools would pay teachers handsomely for referring their friends and family to become teachers on those platforms as well. By making clear at which stage of their teaching journey a referral reward would be paid out, advocates could reliably plan their monthly income around their referrals. Over the years, some teachers stopped teaching altogether and generated their income from referring teachers exclusively. A good referral system can do that.

The second significant advocate benefit is reputational. If you show to your peers and your community that you use an expert tool to do your job expertly, what will your peers think you are? An expert, of course. By embedding the referral link into helpful content, many advocates show that they know what they are talking about, and then benefit from people following their advice by clicking the link.

The third advocate benefit is related to empowerment. It feels good to help other people, and it forges relationships. I am still in contact with many people who have referred business- and programming-related tools to me. This is not because I saved a few dollars, but because I received their advice and guidance, with the referral being a small part of it. They helped me understand that things can be done better. They helped me grow. I have since done the same, and I feel good knowing that my advice and suggestions are enabling other founders and developers to become better at their craft.

What Benefits Does a Referral System Provide to New Users?

It is quite likely that your onboarding is not as effective as you think. Prospective customers come to your landing page, some of them sign up, some of those might use your product, and a few of them will eventually subscribe.

Referrals are the duct-tape for your leaky funnel. A referred user comes to your product with the expectation that if the advocate can use it, so can they. There are a few incentives to sticking with a service that comes referred: trust in a supposedly good product, the expectation of savings down the line, and, most importantly, knowing that there is a real person somewhere that they can ask if they run into trouble: the advocate.

How Does a Referral System Help Your Business?

The relationship between the new user and the referring advocate is important, and shouldn’t be underestimated. In fact, it can be incredibly useful to your business.

Let’s look at the example of Chinese online schools again. They built a referral system that would show the advocate exactly where their referred teachers were along their journey. And then, they would allow advocates and new users to communicate through their platform. They encouraged advocates to help and guide new teachers through their journey.

Essentially, they crowdsourced their onboarding to existing customers. If you build a referral system where your experienced customers help the new users get up to speed, you can significantly reduce the workload of your customer service department. If you add a referral system on top of a mature product, you can expect the most basic questions to be answered by your advocates before they even make it into your customer service tickets.

We built such a journey-tracking component into the FeedbackPanda referral system. Advocates could see if their referred users accomplished certain milestones, and where they struggled.

This kind of outsourcing also creates a permanent feedback loop about the onboarding process and first-usage problems. Right after we released the FeedbackPanda referral system, our advocates started reaching out, telling us about where they had issues helping their referred users make sense of the product. This allowed us to either change the product quickly or provide better instructions through our knowledge base.

We also noticed higher retention rates in both cohorts, advocates and referred users. We attribute this to the long-term reward for the new users, and the easy-to-reach threshold of three referrals per month to perpetually have a free subscription for advocates.

One of the most significant benefits of having this system turned out to be the added insight we suddenly had into our customer base. With unique referral codes, we could track where our new customers were most receptive to marketing, and who among our customers would most actively share their link in which social networks. That made the previously opaque word-of-mouth marketing much more measurable.

We toyed around with the idea of creating different referral codes per social network but never followed through. The idea was to make the tracking even more manageable, but we decided against that. This becomes particularly cumbersome when you don’t create random referral codes but use human-readable ones, like with the name of the customer or their business.

What Risks Are There?

While referral systems are usually opt-in and can be ignored entirely, there are several questions you might want to ask yourself before offering a referral system.

The Risk of Fizzling Out: Is Your Service Shareable?

Referral systems don’t work for all products or services. In fact, there are many reasons why a referral system wouldn’t work too well. Let’s look into a few of them.

A referral system might fizzle out if:

• My Precious: Customers would lose their edge if they shared your service. Think of a marketer finding a tool that allows them to capture new hidden markets. Would they share their secret weapon with other marketers?
• Training Wheels: Customers would admit to being a beginner if they shared your service. People rarely admit they’re not good at something.
• Empty Pockets: Customers would admit needing the savings from the referral system. We don’t like admitting not to be able to afford something. If sharing a link might make other people think we’re cheap, we’d instead not share it.

Conversely, a referral system will work best if:

• Network Effects: Customers stand to gain something from another user joining the service.
• Reputation Gain: Customers can show that they are experts in their industry to their peers.
• Way Above the Bottom Line: Customers can show that they are doing well using your service.

Ryan Kulp wrote a blog post on the dimension of shareability that I highly recommend reading.

For FeedbackPanda, we had a powerful network effect built into the product: our customers could share their feedback templates with each other. That meant that every new customer might bring fresh content that the existing customers could use immediately. Inviting a new customer to the website was a way to do less work eventually for our customers, so they had no problem sharing the service. Our service also showed them in a professional light: teachers who use FeedbackPanda care about their students and are good at what they do. And finally, we could disregard the financial aspect, as almost all teachers are well aware of their precarious financial situation and don’t mind sharing bargains with each other.

The Risk of Feeling Icky: Will It Impact Genuine Mouth-To-Mouth Marketing?

Once you’re sure that your product is inherently shareable, think about how it might change the conversations that already happen organically. Often, a recommendation that came naturally can feel forced or sneaky.

If you can live with the fact that some of your prospects might be turned off by incentivized sharing, then go right ahead. But if you have a very sensitive niche, be very careful how and how much you recommend sharing referral links.

And here’s the thing about sufficiently large groups of people: you won’t be able to please everyone. There will always be people who are seemingly allergic to even the slightest sign of self-interest in their communities. We had a few teachers complain about the referral system cheapening the genuine messages of their peers. These purists will cry out whenever they sense the slightest trace of rewarded behavior. They will call their industrious peers “shills,” accuse them of being “paid for this,” and generally be very vocal about marketing that they think has no place in their community.

If your system works, ignore those people. They are not your audience. Talk to the people who listen. Enable your most excited customers to spread the word and make a little bit extra on the side. If your service is outstanding, people will appreciate it.

We made sure to tell our customers very clearly not to turn this into a marketing frenzy. We asked them never only to post the link, but always to add some helpful text or explanation of our product. We asked them to respect the rules of their communities regarding product placement and advertisement. In consequence, we received very few complaints from our customers about upsetting their communities.

The Risk of Speaking Too Soon: Is It the Right Time to Add a Referral System?

Earlier, we saw that sharing a referral link is always a risk to your advocate’s reputation. What if they refer something of low quality? What if they only share it because they can make a cheap buck? These valid questions appear whenever a prospect is confronted with a referral link.

The best way to alleviate those fears is by providing a reliable product that allows for a great initial experience.

And that is not your MVP. Usually, this happens much later in the lifetime of your business. In the Survival Stage, you try to find a repeatable business model and build features into your product that make it better for your customer. This phase is notorious for lots of trial and error. Lots of things change. It’s definitely not a good time to have customers onboard other customers, only to find what worked for them not working anymore.

Best wait until your product is stable enough for your user-facing documentation not to change every week. That allows your advocates to support their referred prospects more reliably.

Setting up a Referral System

Let’s walk through the two phases of adding a referral system to your service: setting it up and making your customers use it.

Technical Choices

Like most things SaaS, you can either use an existing solution or build your own. I built the FeedbackPanda referral system myself because I wanted to create a very custom system. In retrospect, this probably would not have been necessary. Services like ReferralCandy, ReferralRock, SaaSquatch, or Ambassador offer tools that are easy to integrate and allow you to manage not just referrals, but also affiliates and partnerships. If you want to use a product built by two indie hackers for other indie hackers, take a look at Rewardful.

In our case, we only wanted a simple system that was perfect for our customers. This turned out to be quite some work. Not only did I need to implement all the server-side logic to assign subscription extensions correctly and discounted plans, but I also found myself fixing a large number of frontend-related issues that stemmed from the tracking technology needed to tell who referred whom reliably. If you prefer to spend a few weeks building that instead of useful features, go for it. If you’d instead focus on your strengths, use one of the services that already solved all these problems.

If you use a third-party service, make sure you like the way they integrate. Some use widgets, and some offer APIs. Work with a system that fits with your long-term strategy.

When it comes to complexity, take a staggered approach. Your referral system, just like your product, doesn’t need to be completely finished to be useful. For example, we initially released the FeedbackPanda referral system to track referrals, but not show the current stage of product adoption to the advocate. It was more critical for us to gather feedback on the overall system than providing every feature immediately.

The most important technical aspect of the system is that it reliably tracks who refers whom. That’s all you need to start seeing the benefits.

Encouraging Your Customers to Become Advocates

We launched our referral system publicly, at the same time as we increased our prices. That allowed us to announce both changes at the same time, in tandem. With a month to go, we started reaching out to our customers through an Intercom message and the weekly newsletter, explaining what we’d be introducing and clearly stating the value proposition of using the referral system.

We framed it as a way to help other teachers become better teachers while saving a few dollars. That built sufficiently high anticipation that on the day of us releasing the referral system: there was a noticeable wave of customers sharing their referral links. That prompted a lot of their peers to become interested in what FeedbackPanda was, and those questions summoned our evangelist customers to share their testimonials. People were ready to do our marketing for us, and they did.

In your communication leading up to and after the launch, draw parallels with other referral systems they might already know. It is a trust transfer opportunity that can make a meaningful difference in the adoption of your system.

As our customers already knew a similar referral system from the school they taught for, and might even have become a teacher through going through such a referral process, hinting at the similarities showed them that it was in their interest to participate.

We made sure to publish clear step-by-step descriptions of how to find your referral link and how to make it interesting for prospective customers in several knowledge base articles. We used screenshots and video tutorials to make it as easy as possible to understand. We also added a very prominent interface element into the dashboard of our application to make sure that the concept of referring customers was exposed to our customer’s field of vision every single day.

We chose that kind of communication over active recruiting. We didn’t reach out to people directly, but we put it in front of them wherever we could. A referral system should be optional, and only if a customer really wants to support the service should they do it. We wanted our advocates to refer new customers because they found it the right thing to do.

That doesn’t mean you can’t reach out to specific customers. In fact, active recruiting is an excellent idea in tribal niches. We talked to a limited number of our most socially active customers and recommended that they’d try out the system to see if this would give them more insight into their reach. That approach netted us a few excellent placements of links on niche-specific blogs.

A Few Thoughts Regarding Tracking and Privacy

When it comes to tracking, I recommend you capture the whole journey if possible, for maximum insight. Track the channels where your referrals happen and who is most prolific at putting their referral link into those channels. These are opportunities for partnerships that go beyond sharing a link. You can quickly find an influencer that would like you to sponsor their channels.

If you’re planning to have your advocates onboard your new customers, make sure you restrict their communication in a way that keeps personally identifiable information from leaking. Don’t share email addresses unless your users explicitly agree with such a practice. The best case would be to use an in-app messaging channel that allows you to control who talks to whom.

---

Remember that not every business can benefit from adding a referral system. It might even make the business look greedy when low-shareability products are pushed through a referral system. Referral systems are meant to support your marketing with an incentivized method of getting new users to try out your product. Make sure your product is shareable, your rewards incentivize your customers, and you provide plenty of information on how to they can communicate the benefits to prospective customers.

That’s how you build a fantastic referral system.