7-Eleven · 7NOW App
Slurpee Spark: A Collaborative Evaluation Tool That Activates Personas for Product Decisions
A real-time web application that helps 7NOW product and design teams evaluate initiatives against shared user personas and business criteria — turning static research into actionable, aligned decisions.
My Role
Business Strategy, Retail & eCommerce Research, Product Scoping
Team
Jenny Lian (Product Management), Nam Nguyen (IT / Engineering)
Context
Design Studio for Business — SMU MADI Program
Stack & Methods
User interviews, co-creation workshops, iterative prototyping, AI-assisted synthesis
Designed for 7NOW teams to move personas out of storage folders and into everyday product decisions.
At a glance
The Challenge
7NOW had five user personas, but they were stored away and not driving product decisions.
The Approach
Interviews with 6 team members, iterative prototyping with real users, and AI-assisted synthesis.
The Outcome
A real-time evaluation tool with 5 standardized factors, tested with 10 cross-functional participants.
How might we
How might we activate persona research findings to impact internal design and product decisions for the 7NOW app?
01 — The Challenge
7NOW had invested in developing five official user personas through interviews with 20+ users and a cross-team co-creation workshop. But after the presentation, the personas were stored in a Teams folder and never revisited. Product, design, and research teams were misaligned — research findings were not tied to business KPIs, and there was no standardized way to integrate personas into the product development lifecycle.
02 — The Strategy
We designed a shared decision-making framework where user personas are directly connected to measurable business outcomes. The tool needed to feel lightweight for facilitators while creating enough structure to produce consistent, reusable outputs.
Step 1
Research-first: understand the real pain points
Conducted 6 interviews with 7NOW managers, designers, researchers, and product managers to identify why personas were not being used and what would make them actionable.
Step 2
Define a standardized 5-factor evaluation framework
Co-created with the 7NOW team five key evaluation dimensions: User Personas, Business Impact, User Experience, Team, and Resources. Each factor includes reflection prompts to guide discussion.
Step 3
Prototype iteratively with real users
Built three prototypes in succession — a static scorecard, a digital version, and finally a real-time collaborative session tool — testing each with 10 cross-functional participants and refining based on feedback.
Step 4
Integrate AI-assisted synthesis
Added Gemini-powered insights that summarize session data, surface alignment gaps, and generate discussion questions — turning individual scores into team conversation starters.
03 — The Product
Slurpee Spark is a real-time web application where teams create evaluation sessions, select relevant personas, score initiatives across five factors, and receive AI-generated insights. The interface is designed for facilitation — the host controls the flow while participants join with a 6-digit code.
04 — The Results
The final prototype — Slurpee Spark — was tested with 7NOW product, design, and research teams and validated as a practical tool for aligning initiative evaluation.
Business Impact
Slurpee Spark transforms how 7NOW teams use persona research — from static documents to active decision-making inputs.
- →Replaced ad-hoc workshop debates with a repeatable, factor-by-factor evaluation process.
- →Created a shared language between research, design, and product through standardized criteria.
- →Reduced misalignment by surfacing where teams agree and disagree before committing to initiatives.
- →Produced reusable session outputs (scores, sticky notes, AI insights) for documentation and retrospectives.
05 — Key Learnings
Listen to users and iterate rapidly
The first prototype — a static scorecard — failed because testers did not see how it connected to personas. We pivoted to a real-time collaborative tool after just one round of feedback.
Structure creates alignment, but flexibility preserves collaboration
Teams needed fixed evaluation factors to keep sessions comparable, but they also needed open-ended sticky notes and reflection prompts to capture nuance.
AI insights work best when grounded in explicit data
Gemini-generated recommendations became useful only after we tied them to structured factor scores and session context — not when they were generic.
Cross-functional teams need conversation, not just scores
The most valued feature was not the final score — it was the composite view that sparked group discussion and surfaced hidden assumptions.
Need a collaborative evaluation framework for your product team?
I design research-to-action systems that turn user insights into aligned, measurable product decisions.