We partnered with the GetGFTD team to build the behavioral foundation for a high-stakes redesign of their gifting-first payment platform. From heuristic diagnosis through usability validation and post-launch measurement, our work surfaced the structural friction blocking early retention and translated those findings into decisions the team could act on with confidence.
GetGFTD is a strength-based gifting platform built around sending and receiving cash gifts through social connections. Ahead of a major redesign, the team needed to understand exactly what was driving disengagement, not at a surface level, but structurally. The existing product had 14 top-level navigation items split across three separate menu systems, a non-core feed feature diluting the gifting experience, and onboarding flows fragmented enough to cause significant drop-off before users reached the product's core value. With a redesign already scoped, the risk wasn't building the wrong product from scratch, it was rebuilding the same structural problems with better visual design.
Outcome: Week-one retention improved from 40% to 76% following a three-phase behavioral research and redesign strategy.
Research across three phases surfaced four signals that shaped the redesign strategy:
- Navigation architecture was the primary conversion blocker, not the value proposition. The app split navigation across a hamburger menu (14 items), a bottom bar, and a right-side sticky section. In moderated testing, 100% of participants misfired on wishlist navigation, and 67% rated the task 1 or 2 out of 5 for ease. These were predominantly tech-comfortable users (67% rated their technology comfort 4 or 5 out of 5). The problem wasn't user capability; it was an information architecture that made the core action functionally invisible to people who use apps every day.
- A non-core feature was actively competing with revenue behavior. The feed created complexity without supporting GetGFTD's primary function: sending and receiving gifts. Because gifting is the platform's revenue mechanism, every interaction the feed diverted was a transaction that didn't happen. Removing it wasn't a feature cut; it was a strategic prioritization that let the core gifting loop breathe.
- The entry-point problem was structural, not experiential. 56% of new users attempted to log in instead of sign up, a default screen hierarchy issue, not product confusion. Onboarding satisfaction was 100% at 4 or 5 out of 5 among participants who reached the correct screen, confirming the flow itself was sound. One participant even articulated the standard: "Most apps automatically start with sign-up." The fix required zero design effort, but left unaddressed it would have compounded into significant acquisition loss at scale.
- Without a measurement architecture, the redesign couldn't learn from itself. The team had no defined Mixpanel event schema before the engagement, which meant post-launch behavioral changes couldn't be attributed, validated, or acted on. Studio Marelle designed a production-grade schema covering onboarding, wishlist, social, bank linking, and gifting checkout with event-level properties (amount, recipient_type, payment_method_type, status) and cross-platform iOS/Android parity, before the redesign launched. That infrastructure turned a one-time intervention into an ongoing feedback loop, and made it possible to identify the gifting checkout gap as the next clear priority the moment the data became available.
We engaged across the full pre-redesign diagnostic and validation cycle, working directly with the founder, design, and engineering teams:
- Heuristic Evaluation: systematic review of sign-up, navigation, wishlist, gifting, and homepage flows against usability standards; severity-ranked findings with implementation roadmap
- Moderated Usability Testing: 9 remote sessions across onboarding, wishlist creation, and gift-sending flows; task completion metrics, error rates, confidence ratings, and direct behavioral evidence
- Research Synthesis & Decision Mapping: prioritized findings translated into specific design and engineering decisions, with effort/impact framing for the redesign team
- Mixpanel Analytics Schema Definition: production-ready event tracking specification delivered to the engineering team, covering onboarding, wishlist, social, bank linking, and gifting checkout with event-level properties, identity management, and iOS/Android parity requirements; defined 11 priority events required before public launch
- Pre-Launch UI & Bug Audit: systematic QA review of the live TestFlight build identifying P1 issues (including a critical wishlist-save bug and UI inconsistencies affecting trust) with severity ratings and launch-blocking recommendations
- Post-Launch Behavioral Analysis: pre/post Mixpanel comparison across retention, daily active users, social engagement, and gifting conversion to validate redesign impact and surface the next layer of opportunity
GetGFTD launched the redesign with a clear behavioral foundation and the infrastructure to measure what happened next. Week-one retention improved from roughly 40% to 76%, a 90% relative gain, suggesting the structural changes to navigation, onboarding hierarchy, and flow consolidation meaningfully changed how new users experienced the product for the first time. Daily active users stabilized at a higher baseline, and social engagement (friend requests) increased approximately 3x at peak, a meaningful signal for a platform where gifting depends on social connection density.
The Mixpanel schema meant the team didn't have to wait for problems to become visible in product reviews or support tickets; behavioral patterns surfaced immediately. The post-launch analysis identified gifting checkout conversion as the next active friction point, giving the team a specific, data-grounded priority rather than a general sense that something was off. The work gave GetGFTD not just a stronger redesign, but a durable feedback loop that makes each subsequent iteration more informed than the last.
