What if someone told you that you could build an acquisition channel that could make up >20% of your new user traffic at a customer acquisition cost of 50%-70% less than your paid channels and would require zero maintenance after optimization?
I’ve owned Referrals at Lyft for both driver and rider acquisition and built the referrals product at Opencare, a Series A company. Through my experience and my conversations with other product owners, referrals have similar metric benchmarks across companies. Here is a simple model that has average benchmarks to help you size the opportunity of your referrals program.
Now, let’s get to the tactics.
Top of Funnel - Marketing your Referrals Program
Optimizing the top of the funnel (or adding entry points) is the most impactful. If you go back to the model, you can see that “% of people who view the referrals screen” has the highest coefficient.
I have a simple framework for brainstorming entry points:
- Which parts of the product do users experience peak happiness?
- How do users refer others to the product right now?
Let’s use Lyft as an example.
1. Which parts of Lyft do users experience peak happiness?
- After signup, inside the ride, post the ride.
- When a user is in any of these contexts, we show either an email or a product prompt, like the one below.
2. How do users refer others to Lyft right now?
- Most referrals happen in the moment, specifically at a restaurant or bar on a Saturday night. We also knew that one of the top customer support complaints about referrals is that users can’t find their referral link. Given this finding, we tested different copy in the side menu. Eventually, “Get $20 free” was the best performing variation. This was one of our most impactful tests.
- Additionally, we would try to nudge them in the right moment with the right message. During these “peak referral times,” we would surface an entry point with the copy, “With friends? Refer….”
Middle of the Funnel - Optimizing Your Referrals Screen
The referrals screen has three elements: attention tactics, who to refer, and how to refer.
- Attention Tactics: Incentive vs Non-Incentive
When growth teams think about referrals, they often immediately think about the incentive. If we widen our lens, we can re-frame the benefit of the incentive, which is to attract attention. How else might you attract attention? With other non-monetary tactics.
Incentive
- Instead of $ for “Give $, Get $”, use %-off discount → At Lyft, we discovered that %-off converts better and is more cost-efficient.
- Find the optimal amount for both the ‘give’ and the ‘get’, and build a cost curve → Test different dollar increments, e.g. $10 vs $20 vs $30.
- Get rid of the ‘get’ → Users’ altruism is the main motivation to refer.
- Give users options on how much they should ‘give’ and ‘get’ → Users enjoy options, and this is a great engagement tool.
- Can’t give store credit? Use gift cards → If you don’t have a functionality to reward store credit, give out branded gift cards, e.g. Amazon.
Non-Incentive
- Create scarcity → Tell users that they only have 5 rewards to give out to friends.
- Explain why referring is beneficial → There is evidence that when you explain why you’re asking, more people are likely to do the request.
- Exclusive club → If users refer X number of people, they are accepted to a club.
- Get users to share an image → Instead of getting users to send a referral link, which feels transactional, ask users to share images or infographics instead.
- Who to Refer
Knowing who to refer is the hardest part about referrals. To make it easier, suggest specific people. To do this, you need to understand who the referees are through qualitative and quantitative research.
- Tweak copy → At Opencare, we discovered that 50% of referees are the users’ partners. We then tweaked the copy in our referral screen, specifically the text inside the email input box. Just with this slight tweak, we saw a significant increase in the number of email addresses inputted.
- Identify contacts who are already using the product → If you ask for users’ consent, you can identify current users in people’s contact lists, which also means that you’re filtering those who aren’t using the product. Strava is a great example of this.
- Utilize the Zeigarnik effect → This effect happens when we do not complete a task that asks us to fill in the blank. This creates mental tension that naturally encourages us to solve it to feel satisfaction. For example, if I was the Growth PM at Strava, I could say, “the last person I biked with was….”
- How to Refer
Last but not least, how to refer, meaning what are the means through which users refer.
- SMS / Text Messaging Apps → The #1 medium is SMS. At Lyft, >80% of referrers share through text message.
- Email → The benefit is tracking, but the disadvantage is user experience, since users cannot recall their contact’s email address.
- Referral Link → Majority of referrals happen in the moment, and referrers are rushing to find their link.
- Social media apps → Don’t waste your effort. Only 2-3% of users share on social media, and those who click on the link almost always bounce off the landing page. When I tested this at Opencare, only 1-5% of viewers of the landing page converted compared to the typical 40% conversion rate.
Bottom of the Funnel - Landing Page
Like all landing pages, personalizing the page will increase conversion rate.
- Add the referrer's first name in the headline → The more personal variables added, the better, but ensure you are GDPR-compliant.
- Show the referred product item, not the company → Handy, for example, directs their landing page to the profile of the contractor. Many e-commerce stores also direct their landing page to the specific referred product.
- Personalize info to the referee → At Lyft, we implemented an earnings calculator for driver referees to see how much they could earn in their region.
Referrals is just a simple funnel that, when done well, could scale to be one of your biggest channels for growth. Just start from the top of the funnel and work your way down. Good luck!
