When a user tries to read an article on nytimes.com, the “regi-wall” prompts them to create a free account to keep reading. After they register, we take them through a series of onboarding steps before sending them back to their article. I maintain this registration flow with my team, and have tested and iterated on it extensively.

I have learned that the most successful interventions achieve our business goals by speaking to a user need in the right moment. Below you will find some documentation of the flow, as well as some of our most informative tests from the past few years.

Role
UX Design, Visual Design, Research Facilitation

Collaborators
Product, Engineering, Data, Research, Copy Writing

Magic Linking Iteration

Additional testing reassured users that they we would automatically log them in once they download the app. Messaging improved intent to download the app by 2%. Functionality + messaging resulted in 92% increase in users immediately populating newsreader apps. 

Article interest iteration

An iteration of the above interest test looked at the impact of using articles to demonstrate interests. My hypothesis was that interests like “science” or “world news” can be too abstract and aspirational, and that users might be more able and willing to demonstrate interest in individual pieces of content.

This was a painted door test at first (not hooked up to anything on the backend) to determine whether we would have a higher take rate.

While I think this hypothesis has merit, and it has informed later work, this was ultimately not the right moment for this type of experience. I believe it was too much content for users to engage with in a moment of friction. Take rate did not improve compared with control.

Products in Get Started (2021)

We have been looking for ways to gather signal on which users might be interested in our other digital products (Cooking, Games, Wirecutter, Audio).

I designed a simple step in onboarding that would allow users to select content types they like, including ones that mapped to our core digital offerings. This was a preliminary test to see how many folks would demonstrate an interest in media type. In the future, we may return to this step, with more focused visuals expressing the Cooking, Games, and Wirecutter brands.

User Testing

While exploring this concept, we did some user testing with wireframe prototypes. This helped us determine which content types to focus on, as well as some of the framing. We learned that the copy “tell us about yourself” made users feel as if this was for The Times, not for them. They worried we were just trying to track them

Get Started quickly + dark mode (2021)

Onboarding, but make it faster. The simplicity of the personalization and signup flow was appealing to users, though some of the interactions needed fine tuning. I still think this idea has merit but it didn’t end up making it to the top of our priorities.

The “dark mode” aesthetic was extremely appealing to users. Many noted that it helped them understand that this was an interstitial moment and that they would get back to their story after it ended. Based on this insight, we did a simple A/B test of light to dark mode. The dark mode version did not beat control. However, we have a hypothesis that it lost because the steps before onboarding couldn’t yet be shown in dark mode. It’s in our backlog to explore this further.

Election Onboarding (2020)

I wanted to explore ways to make the onboarding experience feel more like a useful tool for users signing up on election articles. In user testing we saw that the flow felt too long and arduous to folks. Some of the ideas were appealing (follow specific storylines) but it wasn’t exciting enough to demand prioritization in our roadmap, especially given that allowing users to follow a storyline would be a new capability to build out.