Asking Less to Get More
How I proposed replacing a data-collection gauntlet with a trust-first system that actually activated users.
The product took before it gave.
Before a new user could do anything useful, the product asked for everything up front: identity, verification, a financial instrument, regulatory sign-offs. Each request had been added by a team trying to boost their own product's readiness, but no one owned the total the user actually faced.
The original compliance floor was legitimate. What piled on top of it over the years was clutter no one owned, and my onboarding team had to reconcile it.
Our hypothesis was that a maximum signup gauntlet, asking for information with no context for why it mattered, was driving mid-signup account abandonment. Users had no reason yet to trust us with an SSN or a bank link, so we were asking for the highest-stakes information before we'd earned any of it.
For users who pushed through the gauntlet, our second hypothesis was different. Because the flow was optimized to get people through it, not to teach them anything, they never built a working mental model of the product. They were smashing confirm buttons to reach the one task that brought them there, not grokking what else was available or how it fit together.
Research even made this second problem look smaller than it was. Survey and brand-recognition scores were fine. Put those same users in front of real tasks, and features plainly visible on screen failed to register as available to them.
Users weren't confused about what PayPal is. They were confused about what PayPal could do for them right now, in this moment, with this account.
The flow was optimized for completion over understanding. It produced users who could finish signup but couldn't use the product, and whatever mental model they formed decayed fast.
Only ask for information when you're giving something back.
To test whether context mattered, we asked for the same things (offers, savings, direct deposit, notifications) in two places. When we asked cold at signup, conversion was low. When we asked in context, right after a warm moment of value, the same ask converted far better. The data collection stayed identical. What changed was the user's reason to say yes.
Push opt-in was the most valuable attention we could earn, and the clearest example of asking for too much, too soon. At signup no relationship existed and no value had been delivered, so users had little reason to say yes. iOS only lets you ask once, so a decline ended the conversation for good.
We changed the conversation. Anchor the ask to a moment the user already cared about, a package they were waiting on, and use a soft proxy first, preserving the single real iOS prompt for a high-intent moment.
The bigger bet: ask for nothing up front.
At the time you couldn't open an account without adding at least one financial instrument, a bank, debit, or credit card. Customers who added a second instrument later showed higher MAU, fewer declines, and fewer negative balances, and that was the signal. To be clear, this was correlational. Engaged people add instruments, but adding one doesn't by itself make you engaged. The hypothesis was strong anyway. Require no instrument up front, ask for one only at first product use, and select for higher-quality accounts the way the notification move had.
The P2P team wouldn't test it. The understandable concern that pushing the instrument step into the send-money flow would add friction at the conversion moment and cost completions blocked live testing. Our qualitative work pointed the same way, though, and the notification result was a strong analog. We had the evidence to make the case. What we lacked was a team willing to own the risk.
When you can prove it works but can't ship it, the problem is organizational.
Name the signup platform a product-infrastructure risk, loudly and early, in business terms. Working around it was pragmatic short-term and limiting over time. Every cycle spent designing around the monolith was a cycle not spent extending the system's reach.
When you can prove a design works but can't ship it, the remaining problem lives in the organization. Treat it that way. Even so, progressive reinforcement, trust-gated credit, capability-aware surfaces, and the notification-proxy architecture carried into later design thinking. The principles outlasted the org changes that blocked them.