Onboarding and Activation
I was one of two mobile designers on the growth team. I introduced and led our team to use Lottie, a framework of scaling animations on different devices launched by AirBnB.
When I joined the growth team, we were in the habit of adding lots of animation, delightful moments in order to encourage people to onboard into our app. At the time, we were using primarily GIFs for animations which has compression issues (up to 256 colors only), and did not scale for different devices.
I worked with engineering to launch Lottie into our day-to-day workflow. After setting Lottie up on After Effects with Bodymovin’ plug-in, I taught two of our designers how to use AE and export animations into .json files for the engineering team. We continued to use Lottie for animations after I left the team. I worked with engineers to optimize speed.
Evernote is the perfect company to think about growth. Growth for the most part is misunderstood. Usually, when people think of the growth funnel, they think of the marketing funnel, which is the top 2 segments in this upside triangle. Growth touches on all aspects of the funnel. As a designer on this team, every single stage is a user touchpoint. We don’t just care about discovery, but we also care about activation, retention, revenue, and referral. We have to understand the product in a holistic way, because only then can we guide the users to see the value in it.
We start out with an experience map where we look at the actions and mindset that a user might have when they are researching a productivity app. What makes Evernote unique from all other productivity apps out there? From there, we can discover moments in a user's journey where they might feel convinced (or, not convinced) and find opportunities to retain them through the onboarding process. It's important to recognize that onboarding is not just the first experience.
That's where retention comes in. We care a lot about NAU (Newly Active Users), which is a metric we use to measure retention.
A user is considered newly active if they create 1-2 notes in their first session, as well as come back for a 2nd session within their first week.
We've discovered that users who are in this data point tend to come back and become long-term users, and many of these users will be retained and end up converting. Although, it's not a total loss if they do not end up converting because we have also learned Newly Active Users do end up converting at some point in their journey, especially if they find value in our product. At the end of the day, users will pay for a product that they find value in. The problem isn’t getting users to pay us, it’s to get users to find the value so that they are willing to pay us.
Experiment A
We ran an experiment where we asked users about their use case and then showed them relevant note types based on their answers. We introduced a checklist experience that will guide them through their entire onboarding journey, including the first launch experience throughout their entire life cycle with Evernote.
We learned from this experiment that it lifted metrics we didn’t even personalize. People felt like they were getting a personalized experience, and so they decided to proceed and come back.
The result was a 3% activation (NAU) lift across the board, and we saw users coming back the second and third week. We learned from this experiment that it lifted metrics we didn't even personalize. People felt like they were getting a personalized experience, and so they decided to proceed and come back.
Although our checklist experience did not lift retention, it drove a lift in the usage of features. It drove a 60% lift for web clipper (one of our features), and 8% for attachments. 30-40% of users who came through the onboarding engaged with the checklist. We found that users who selected the student use case engaged the most with the checklist.
Experiment B
In onboarding, a lot of the times the flow cannot be too long because of the drop-off problem. Logic says longer flow equates to higher drop off rates. While this is true, because users were engaged from the get-go and saw value in answering all questions, there was very little drop-off despite a longer flow.
For a second experiment, we added 3 questions on top of version one. We expanded and personalized the checklist.
From this experiment, we saw a 5% lift on newly active users, a 3% lift on users coming back the second week, and a 7% lift in week 2 and week 3 in note creates.
It improved engagement across the board and personalizing the checklist led to 7% more users using it. In onboarding, a lot of the time the flow cannot be too long because of drop-off. However, in this case, users were engaged from the get-go and saw the value in answering all of the questions. There was very little drop-off despite a longer flow.
Mobile Experience
