MURAL's sign-ups were down 61% year-over-year. I led the end-to-end redesign of the authentication experience — defining the strategy, aligning stakeholders across engineering, legal, marketing, and analytics, and executing a two-phase redesign that stabilized critical failures before rebuilding the experience from scratch. The result: +6% signup conversion, -33% profiling abandon rate, and first-party data that reshaped how the team approaches onboarding.
By September 2024, MURAL was facing an alarming decline in user acquisition — sign-ups down 26% month-over-month and 61% year-over-year. While multiple teams were mobilizing around the broader growth challenge, my team took ownership of the authentication experience as the highest-leverage surface to address.
Funnel analysis revealed the root cause: only 52% of users who reached the signup page completed it, with the first screen alone responsible for 84% of all drop-off. The problem wasn't traffic or brand awareness — it was friction embedded in a flow built on technical debt, broken edge cases, and no coherent design vision. This was fixable, and the upside was significant.
I reframed the problem early: this wasn't a UI refresh, it was a product problem that needed both a short-term recovery plan and a longer-term redesign. Trying to do both at once would have meant shipping polish on top of a broken foundation. Instead, I defined a phased strategy — stabilize first, then rebuild.
Before any design work started, I aligned stakeholders across engineering, analytics, legal, marketing, and content on goals, constraints, and sequencing. I partnered with the analytics lead to instrument every KPI from the outset, so that every decision we shipped was measurable and defensible. This upfront investment paid off — we had data to back every trade-off we made.
Phasing the delivery. The temptation was to redesign everything at once. I pushed back on that. Phase 1 focused on resolving the critical failures causing immediate abandonment — bugs, broken flows, missing SSO. Phase 2 was the strategic redesign. This sequencing reduced risk and let us ship real improvements fast while setting up a cleaner foundation for the bigger work.
Adding 4 profiling questions despite the friction tradeoff. Marketing and personalization needed first-party data we weren't capturing. We knew adding questions would introduce some friction, but I structured them post-account creation — out of the critical conversion path — and framed them as a benefit to the user. The bet paid off: completion improved by 11% and the data reshaped how the team approached onboarding.
Invisible reCAPTCHA over a visible challenge. We had a bot problem, but a visible CAPTCHA would have added friction at exactly the wrong moment. I drove the decision to implement the invisible version — which required alignment across engineering and legal — because protecting the user experience mattered more than taking the easier technical path.
Renaming "Sign-in" to "Login". A small copy change that required real cross-team buy-in. These are the decisions that only happen when someone is actively thinking about the whole system, not just their own slice of it.
The hardest part of this project wasn't the design — it was the competing priorities pulling in different directions simultaneously. Legal needed full consent coverage before launch. Marketing needed profiling data to improve lead scoring. Engineering needed scope to stay manageable. Users needed less friction. These didn't all point the same way.
Getting alignment on the profiling questions was the clearest example. Marketing wanted them upfront, in the critical conversion path. I pushed back, arguing that placing them post-account creation — after the user had already committed — would protect conversion while still capturing the data. That position required trust from marketing and a clear hypothesis. The results validated the call.
I also introduced something the team hadn't had before: a shared measurement framework defined before design started. This changed the conversation from "do we like this design" to "does this design move the metric" — which made decisions faster, reduced revisits, and gave the team a way to defend design choices to leadership with data rather than opinions. That practice outlasted the project.
Phase 2 also meant holding a high bar for visual and interaction quality across a surface that had been neglected for years. Every screen went through design system alignment, accessibility review, and responsive implementation. Standardizing interaction patterns across the entire flow — from error states to email verification — wasn't just about polish. It reduced cognitive load and built the kind of consistency that makes a product feel trustworthy at exactly the moment users are deciding whether to sign up.
After the April 2025 launch, the hypothesis held: removing friction and sequencing data capture more thoughtfully produced significant gains without major engineering lift.
- Signup conversion: 52.8% → 58.5% (+6%)
- Profiling abandon rate: 4.5% → 3% (−33%)
- Profiling skip rate: 33% → 27% (−18%)
- Profiling completion: 63% → 70% (+11%)
The profiling data also surfaced something we didn't anticipate: 70% of users had never used a visual collaboration platform. That single insight changed how the team talked about MURAL's market position and reshaped the onboarding strategy. I didn't foresee that impact when I was arguing for the feature — which is a reminder that design decisions can create value in directions you don't predict.
What I'd do differently: push for qualitative research earlier. We relied heavily on funnel data to identify where people dropped off, but had limited signal on why. The 84% first-page drop-off told us where the problem was — not what was going through users' minds. We made educated guesses that turned out to be right. Next time, I'd want evidence, not just good instincts.
The thing I'm most proud of isn't the conversion number. It's that this project established a different way of working — where design owns the measurement framework, not just the experience.