Strategy

• Focus on the gen-pop value prop: The features included in the Gold membership aren’t universally appealing. Our competitive interest rate on uninvested cash was by far the most popular and easy-to-understand benefit, so we focused on finding the right way to communicate the value of that benefit in our upsells.

• Meet people in their context: Plug into existing experiences in a way that feels natural and native.

• Make the value tangible: In several of our experiments, we leaned into leveraging earnings projections to help users quickly understand the value of Gold in a concrete way. This required a lot of negotiation with Legal and Compliance, but the results ended up being worth it.

Process

When I joined in the fall of 2023, the growth design team hadn’t formalized processes for hypothesis definition, experiment design, stakeholder management, and experiment documentation. I led a workshop with designers across all the Growth teams and then summarized the takeaways from it into a growth design process playbook. This playbook served as an anchor for teams learning to operate in this model and a key onboarding resource for new designers.

I visualized this process as a circle because the latter steps of experiment analysis, documentation, and socialization typically directly influence the team’s subsequent prioritization decision for the next experiment. In the playbook, I wrote up detailed guidance for each step of the process using my experience from being on the team and learnings from the Growth Design course I took in spring 2024.

At a glance, this looks like a lot of steps, but I’ve found that once a team embraces a strong process it begins to feel like muscle memory to move through it. And of course, not every step is necessary or possible for every project. For example, most of the time we unfortunately wouldn’t have UXR bandwidth to do any lightweight testing ahead of an experiment. Despite having an approvals-heavy culture though, our pace as a growth team was overall fairly fast—we’d typically bring an experiment from conception to implementation in 2-3 weeks.

Experiments

Gold native funding upsell

Previous attempts to integrate a Gold upsell into new user onboarding had yielded some incremental success, but we knew that we could drive a lot more impact in such a high-intent flow.

Our onboarding flow is structured into different sections that all begin with these chapter header screens. The steps are all pretty functional and in-line with industry standards for a brokerage onboarding flow. The team had previously chosen to slot the Gold upsell after the funding flow, since that made it the last step before the user entered the app.

The previous team had boiled down the Gold upsell in onboarding to a more functional choice. Right after a user would make their first deposit, they'd "choose their interest rate." This had been the most successful of the previous iterations, resulting in a 40% lift to Gold adoption in this flow.

In a team brainstorm, I had the idea to find a placement for the Gold upsell within the funding step in onboarding, since this would more contextually present a deposit-related benefit (earning interest) at the moment of making a deposit.

Hypothesis: If we tie the Gold upsell more closely to funding, more users will sign up for Gold.

Variant 1

First, my product designer suggested leveraging the existing placement, but not segmenting the Gold upsell off with a separate chapter header. Since we had seen previous success with the selector design, she suggested repurposing it, but making modifications to make it feel more contextual within this flow.

We arrived at this design for the first variant. The center-aligned text and progress bar make it feel like it's still a part of the funding flow, but it uses the same selector design from the previous experiment. The subheader content also highlights the amount they just deposited.

Variant 2

However, I had been trying to think about the funding flow and Gold upsell holistically as an information experience, and I suspected that post-deposit still wasn't the right moment, even if the design felt more native. We were asking users to opt in to a membership that offered significant deposit-related benefits, like APY and FDIC insurance, but we were doing it after they had already put money into Robinhood.

Funding is also our highest-friction part of onboarding, so more people would see the upsell if we moved it earlier, before the user has to make the often-difficult decision of deciding how much to deposit.

The selector design, however, didn't make sense pre-deposit—you can't ask users to choose an interest rate for money they haven't deposited yet—so I designed this second variant, which reuses existing patterns from the funding flow. The design itself is pretty unremarkable, but I tried to make the content feel contextual by telling users why they're seeing this now ("you're about to make your first deposit on Robinhood—want to take advantage of these deposit-related benefits we have?").