Referral Program

Background

I initiated and led a design sprint with the growth team and identified a referral program as a high growth opportunity to utilize the 80M subscribers we had at the time. I further validated this hypothesis by initiating and planning user research to measure willingness to refer, expectations for rewards, and perceived value of rewards to weigh against cost.
‍
Next, I together with my product manager partner, pitched leads on greenlighting the referral program and sign off on the financials. After successfully getting alignment and approval the project was started.

Team

I drove the project along with the PM, I also worked with UXW, UXR, engineers, finance, analytics, legal, and I managed contractors who did our illustrations and email designs.

Time Frame

From initiating the project, to final hand off to engineering was three months.

Competitive analysis

After we had validated through UXR that users were likely to refer and what incentives motivated them, I then started with a competitive analysis. I looked at popular referral programs, new apps, programs from large established companies, as well as the few within Google. After I had collected a large library, I studied the patterns they chose for the elements we needed, what elements we could add/use, and which were the most effective.

Through this analysis, I learned that the most used and effective sharing option was to quickly copy the share link. Many had fancy share functions and even our engineering team wanted us to use our Google sharing sheet. However, those options, while helpful, were not used as often and created friction for users. Most users want to quickly get the link to use naturally where they communicate, be it text or email, a social media comment, etc. So the more advanced share options, which would allow us to pre-populate a message about the benefits of the Google One subscription, would be moved into a single share button. Lastly, I learned that tracking the progress and rewards motivated users to share and was important to display prominently.

Challenges

Getting alignment on what the core incentive strategy should be, was a big challenge. We had strong voices from our partners in other Google products wanting to explore different incentives but we were able to drive alignment through the UXR results paired with a report we had our finance team put together. This made a sea of opinions focus into a single obvious answer. We had all the data and it made the path clear. However, after doing some additional hands on exploration work, our engineering team hit a blocker on our main incentive we had chosen. The engineering time cost was too large but thanks to our data, we knew what the next best option would be.

I also had to fight to keep the incentive strategy simple for users to understand. There were many advocates with great intentions that proposed additional steps or conditions that would make the referral program harder to understand for users. For example, we had considered a time period a referred user would need to be subscribed before the referring friend would get their reward. Instead I was able to address their concerns by reworking the incentives in addition to the team coming up with other methods of preventing abuse. By holding a strong principle, based on what made referral programs successful, I helped inspire the team to find other solutions.

Success

The referral program has launched as of March 2024 and has already gained some media attention. We will have to wait a few weeks to measure the success in our test countries. While I am proud of the design and it has tested will in usability studies, it is too early to call it a success. Stay tuned!

The referral project has become a foundation for other products to sell Google One and has allowed for more impact to come.

Areas for Improvement

A project is never finished, for this one our aim was to build out the core components and test our hypothesis. There are two areas I would like to address when we iterate on this next.

1. The landing page. The landing page is very static and plainly states facts about the subscription. Users coming here will most likely not know Google One nor be out of storage. Instead, the landing page should explain what Google One is and visualize the benefits outside of storage. Since there are also other pages in our growth strategy that depend in landing pages, I initiated this project and there is more to come here.

2. Leveraging the friend more. Social proof is strong but even stronger when the connection is a friend. This is why referrals are 4x more likely to convert. I would have liked to use the name and/or image of the friend in the landing page and communications, but they introduced vectors for harm/abuse which we didn't have scope to address in this initial version but I believe are worth the increase in impact if the project proves out.