Turning Behavior Into Actionable Insight: AI-Powered Behavior Tracking Tool for Education Teams

ABC does-it helps special education teams turn raw behavior data into clear, actionable insights. Instead of spending hours interpreting unstructured ABC notes, educators can systematically document environmental factors, identify behavior patterns, and develop targeted interventions that lead to faster, more informed decisions and better resource allocation.

My role:

Product Designer

Duration:

12 days

Industry:

Education, B2B SaaS

Deliverables

User Interviews & Surveys

Stakeholder Map

Competitive Analysis

AI-driven Prototypes

User Testing

SUS Score Reporting

Storyboarding

92%

task success rate

$1.2K

estimated recovered time value per teacher, per year

88/100

"excellent" System Usability Scale (SUS) Score

Problem Statement

Over 70% of K-12 classrooms rely on paper-based ABC data collection, causing inconsistent interpretation and delays in behavioral insights for education teams, leading to slower, less effective interventions for students.

Project Overview

Design Sprint Ideation

Over a focused 12-day design sprint, I consolidated research, ideation, and prototyping into a single, structured process to rapidly shape the ABC Data Collection Tool. Informed by The Handbook of Human-Centered Design Methods, the sprint established user alignment, mapped behavior-management challenges to design opportunities, and produced a validated prototype ready for testing.

12 day

design sprint

method

Looking

Observing the Human Experience

ETHNOGRAPHIC RESEARCH
Interviews & Surveys

EVALUATIVE RESEARCH
Heuristic Review
System Usability Scale

Understanding

Analyzing Challenges and Opportunities

PEOPLE & SYSTEM
Stakeholder Map

PROBLEM FRAMING
Abstraction Laddering

Making

Envisioning Future Possibilities

MODELING & PROTOTYPING
Storyboarding
AI Prototyping

DESIGN RATIONALE
Concept Posters

Business Rationale: Objectives & Goals

Understanding the Why

Long-term business objective:

Evolve ABC Does-it into a universal behavior tracking platform, expanding beyond education to serve broader contexts. This shift increases total addressable market (TAM) and positions the product as a scalable tool for understanding and improving human behavior.

Business goals:

Efficiency & Accuracy → Reduce average data entry time while improving data consistency and clarity with responsive design

AI-Assisted Insight → Use machine learning to highlight trends, flag risks, and generate actionable recommendations that support education teams

District Value → Provide administrators with time-savings and compliance metrics that translate into measurable cost savings

Adoption & Usability → Deliver an intuitive, calm user experience that aligns with real classroom conditions and anticipates teacher workflows

The Solution

Connecting Data Within a Tiered Team System

Guided data collection

I broke down the data import process into manageable steps, ensuring each step comprehensively addresses each subsectionof the behavior import journey.

Design for all devices

Classrooms and districts vary in technology access. Designing for all screen sizes ensures every educator can work seamlessly, wherever they are.

Adminstrative data visualization

Focused data visualization helps admins spot patterns instantly, building day-to-day confidence and enabling faster, more informed decisions.

Clear visibility & comprehension

A clear, calm visual identity reinforces hierarchy and balance, creating an interface that feels structured, approachable, and easy on the eyes throughout the user journey.

Phase 1: User collecting the data



Project Overview

Business Rationale

The Solution

Subjective inputs and manual synthesis efforts slow meaningful insights

Phase 1: Collecting Data

Phase 2: Synthesis & Insights

Phase 3: Planning & Implementation

Define System Strategy

Stakeholder map = understanding the system

Storyboarding the data journey

Design Decisions

Decision 1: design for day to day import

Decision 2: drill down, zoom-in, zoom-out

Decision 3: AI-assisted decision making

Detour! Testing our assumptions

Where we landed