With the ever increasing costs of acquiring users (either to a game, an app or a retail establishment), the ability to acquire users by referrals becomes increasingly important engine for growth. Jason Bosinoff, an engineer at Airbnb, one of the fastest growing tech companies around, recently posted about how it built its successful referral program.
Not only does a referral program help you get users, it generates users who usually have a very high lifetime value. This happens because word of mouth is a very directed user acquisition channel, people will refer a product or game only to friends they believe are likely to value the product.
Airbnb’s referral program is pretty straightforward, with both the sender and recipient getting $25 travel credit when the invited user completes their first trip (or $75 if the recipient hosts). One of Airbnb’s keys to success is that users can send and accept referrals on all platforms (web, iOS and Android), a lesson many game companies would be smart to replicate.
According to Bosinoff’s post, there were five steps to creating the successful referral program:
- Know what success looks like. The first step is to define success. What metrics are you trying to impact and what would represent a positive result. Create three cases: good, better and best.
- Measure. Integrate robust analytics into your program so you can constantly be tracking how you are performing versus the success metrics that you have set. You want to be tracking everything on the customer journey from when they first see the prompt to create a referral to when the accepter rates their experience after using your product. Airbnb built a rich logging taxonomy of over twenty user events that happen during the referral invitation and sign up journey. With this tracking in place they could follow an invitation from invite page impressions to referred users’ making bookings or becoming hosts. They could then easily review an metric or view it graphically.
- Test and improve. You then want to test the product on a subset of your user base. This testing both allows you to improve the stability of the referral program and think of additional functionality that could improve metrics. Some functionality that Airbnb added during its testing process included personalized referral codes and landing pages as well as customizing the experience based on what the user clicked to enter the experience.
- Go live. Launch your referral program and compare with the results you targeted in step 1. As your analytics are implemented across the product, you should easily see how you are doing compared with plan.
- Iterate, iterate, iterate. Compare the analytics you are getting with the plan you established in the first step. Where you are trailing your projections, optimize the referral program to overcome the weakness. For example, if you are seeing fewer invitations per user, increase the incentive for using the program. If you are seeing too few senders, increase surfacing of the program. If you are seeing a weak acceptance rate, change the copy of the referral or increase the benefit to the end user.