Building Connections That Stick: Designing BeeBack

A customer loyalty platform that helped small businesses turn casual visitors into regulars.

Role

Team

Focus Areas

Overview

BeeBack started as a simple idea: helping local businesses reward loyal customers in a digital-first world.
As physical punch cards faded away, BeeBack provided a seamless mobile solution that combined loyalty rewards, communication tools, and analytics into one intuitive platform.

As Creative Director and Co-Founder, I led the design of the brand, product, and entire user experience — ensuring BeeBack felt approachable to business owners who weren’t always tech-savvy.

The Challenge

Small businesses often rely on loyal customers but struggle to build repeat engagement without complex systems or expensive software.
Our challenge was to design a loyalty product that was:

  • Simple enough for a small café to set up in minutes

  • Engaging enough for customers to keep coming back

  • Scalable for larger retail networks with multiple locations

We didn’t just want to gamify loyalty. We wanted to build genuine, long-term relationships between businesses and customers.

We needed to design an experience that:

Discovery & Research

I collaborated with our Product Manager and joined multiple user interviews with instructors and students. While I didn’t directly lead the sessions, I participated in observing user behavior, taking notes, and later reviewed and synthesized the collected data to identify recurring patterns.

I also worked closely with subject matter experts (SMEs), such as professors and math curriculum designers, to ensure the product aligned with both pedagogical best practices and real classroom needs.

Key Insights

  • Cognitive overload often came from unclear steps or lack of context.

  • Students wanted reassurance — real-time feedback, hints, or small wins.

  • Instructors needed visibility into each student’s process, not just outcomes.

  • Consistency in layout and math entry improved confidence for lower-level students.

"Students often defaulted to paper because they didn’t trust the system to help them think through math."

Tom Khulesa, PM, Aktiv Mathematics

The Design Process

01 — Defining the Learning Modes

To support different learning approaches, we designed three modes that scale with student mastery:

  • Scaffolded Mode — Step-by-step guidance for building foundational skills

  • StepWise Mode — Students show reasoning with targeted feedback after each step

  • Final Answer Mode — Standard input for summative assessment and mastery

Each mode was designed to feel consistent yet distinct, with subtle interface and interaction variations that matched cognitive complexity.

02 — Math Input and Feedback

We simplified math input interactions, focusing on clarity and flow:

  • Equation editor supports fractions, radicals, and exponents with natural keyboard entry

  • Graphing tool (powered by Desmos integration) provides real-time visualization

  • Feedback bubbles appear inline to guide corrections, not penalize errors

03 — Instructor Experience

Instructors can now:

  • View step-by-step student reasoning for deeper insight

  • Track class-level analytics and trends

  • Configure assignment settings and difficulty across learning modes

I designed these instructor dashboards to feel less like spreadsheets and more like visual stories of student learning — showing progress, confidence levels, and areas for reteaching.

Collaboration with Experts

Working with subject matter experts was crucial. Together, we reviewed question types, problem structures, and interface patterns to ensure every interaction reflected sound mathematical reasoning.

For example, SMEs helped validate Scaffolded Mode interactions to ensure they matched instructional logic used in real classrooms. This close loop between pedagogy and UX helped make the product truly “teacher-approved.”

Results & Impact

Takeaways

Designing Aktiv Mathematics reminded me that learning tools must respect the way students think; messy, iterative, and nonlinear.

What I learned

  • Balance pedagogical depth with interface simplicity

  • Collaborate deeply with subject matter experts for domain authenticity

  • View error states as part of the learning process, not obstacles to avoid