Rewards Maximizer 🚧
Helping members optimize their credit card spending.
Business context
Credit Karma was acquired by Intuit in late 2020, and this work took place during the early stages of that integration. At the time, teams were navigating a shift in operating model as Intuit’s Customer Driven Innovation (CDI) methodology was introduced alongside Credit Karma’s existing lean, experiment-driven culture.
Credit Karma historically excelled at rapid product iteration—shipping quickly, measuring outcomes, and adjusting based on performance. Intuit brought complementary strengths in structured discovery, customer-centered framing, and decision-making at scale. For this project, I intentionally blended these approaches: using CDI principles to anchor the team in real member problems, while maintaining the speed and learning velocity needed to operate in a highly competitive credit card marketplace.
My goal was not to introduce process for its own sake, but to create enough structure to improve decision quality without slowing execution.
Business goals
At the time, Mint was being ramped down, resulting in an influx of former Mint users into Credit Karma. These users tended to fall into what we defined internally as “Prime” members—individuals with credit scores of 720+, higher incomes, and more established financial profiles.
Credit Karma had historically focused on serving sub-prime and near-prime members, where guidance around approval odds and credit building was most critical. The arrival of Prime members created a strategic tension:
How do we continue to serve our core audience while also building compelling value for a new, higher-expectation segment—without diluting focus or trust?
Leadership alignment was clear that we needed to better engage Prime members, but what that meant in practice—particularly within the credit cards experience—was still undefined.
Business Problem
From tradeline and behavioral data, we observed that Prime members were disproportionately opening credit cards through Credit Karma compared to other product categories. However, conversion alone did not equate to a differentiated or trusted experience.
The opportunity was not simply to surface more credit card offers, but to help Prime members make confident, informed decisions in a category where they already had options and higher expectations around rewards, optimization, and transparency.
User Needs & Gaps (Pre-Discovery)
At kickoff, there were no clearly defined Prime member personas specific to the credit cards space. While Credit Karma had existing personas at a broader level, they did not sufficiently capture how Prime members evaluated, compared, and optimized credit card decisions.
I believed this gap was a risk: without a shared understanding of Prime-specific motivations and tradeoffs, the team would default to assumptions based on sub-prime behaviors or business incentives. I proposed creating lightweight, credit-card-specific personas grounded in real behavior to align design, product, and marketing around the same mental model.
High-Level Prime Profile
Credit score of 720+
Holds multiple existing credit cards
Optimizes for rewards, bonuses, and long-term value
Less motivated by approval odds; more sensitive to relevance and clarity
This profile served as a starting point—not a conclusion—to guide discovery.
My role
I was the sole designer on this project, responsible for setting the experience strategy and design direction end to end.
My scope included:
Defining the problem space and value proposition
Leading cross-functional design sprint workshops
Designing and prototyping concepts
Planning and conducting user research
Synthesizing insights into personas and opportunity areas
Partnering with stakeholders to converge on a shippable direction
Key partners included Product Management, Engineering, Content Design, Marketing, Data Analytics, Business Development, and Legal. While I executed much of the design work directly, my primary responsibility was creating clarity and alignment across these groups as the direction evolved.
Approach
Pre-Kickoff
Before formal kickoff, I reviewed existing research, performance data, and prior experiments related to credit cards and Prime members. Based on these inputs, I proposed a plan that balanced speed with discovery: a time-boxed design sprint followed by iterative concept testing, with product discovery interviews running in parallel.
This structure allowed us to move quickly toward tangible concepts while continuously validating whether we were solving the right problem.
Design Sprint: Divergence with Intent
I facilitated a five-day design sprint to align the cross-functional team and accelerate early learning. Each day focused on a distinct phase of the process:
Aligning on goals, constraints, and success criteria
Generating and critiquing multiple solution directions
Sketching and refining promising ideas
Prototyping the strongest concepts
Testing with members to identify signal, not perfection
The sprint was less about producing polished designs and more about narrowing the solution space based on evidence.
(Winning concepts visualized here)
Problem Discovery in Parallel
In parallel with rapid prototyping, I conducted member interviews focused on understanding real-world credit card decision-making. Rather than asking participants what they wanted, I used a story-based interview approach, anchored around a single prompt:
“Tell me about the last time you opened a credit card.”
This approach emphasized past behavior as a predictor of future behavior and helped surface motivations, tradeoffs, and moments of uncertainty that were not visible in quantitative data alone.
I synthesized interviews into participant snapshots, grouped them into themes, and used those themes to create credit-card-specific Prime personas. I also mapped unmet needs and opportunities using an opportunity solution tree to make tradeoffs explicit and guide prioritization.
(Opportunity solution tree and interview snapshots inserted here)
Converging on the Direction
Across multiple rounds of testing, one direction consistently resonated: helping Prime members maximize rewards across cards they already had, rather than pushing new offers prematurely.
The most challenging decision was not the concept itself, but how to frame it. Different framings tested differently across clarity, perceived value, and trust. To reduce subjectivity, I partnered with analytics to run an A/B test across three variants.
The final direction combined the strongest elements of two approaches. Marketing later named the feature Rewards Maximizer, reflecting both the functional value and the aspirational benefit.
Final Designs & Scope Tradeoffs
For the initial release, we intentionally reduced scope to prioritize clarity and time-to-learn. Non-critical elements—such as dynamic data visualizations and deeper historical breakdowns—were deferred in favor of a simpler, static representation that still delivered core value.
These tradeoffs allowed us to validate demand and comprehension before investing in more complex infrastructure.
(Final mocks and interaction GIFs here)
Outcome
(Insert analytics once finalized)
Early indicators showed increased engagement among Prime members and validated rewards optimization as a meaningful entry point for this segment. Just as importantly, the work created a foundation for future iterations and deeper personalization.
Reflection
This project reinforced the value of balancing speed with intentional discovery—especially during periods of organizational change. The Credit Cards team had strong instincts and deep domain expertise, and my role was often less about generating ideas and more about creating shared understanding, narrowing focus, and helping the team make confident decisions under uncertainty.
Leveraging the collective experience of the team through design sprints, paired with continuous discovery, allowed us to move quickly without losing sight of real member needs..
Leveling Up Other Designers
As part of this work, I shared two presentations with the broader design organization:
An Introduction to Continuous Discovery
Story-Based Interviewing
These sessions helped designers—particularly those newer to discovery—adopt techniques like interview snapshots and opportunity solution trees, improving both the quality of insights and confidence in research-led decision-making..