Nike • Member Experience • Shipped

Designing an AI training companion that turns intent into consistent action.

Role: Product Designer
Timeline: Shipped
Date: 2024
Context: Speculative concept focused on member retention + habit formation
Tools: Figma, Google Docs

Overview

From “I want to train” to a plan that adapts in real life

Nike members often start with a clear goal—run a 5K, lift consistently, return from an injury—but daily variability makes it hard to sustain a plan.

This concept explores an AI training companion inside the Nike ecosystem that personalizes plans, adjusts to recovery and schedule changes, and reduces the decision fatigue that causes drop-off.

Ecosystem map: plan → workouts → recovery → retail loop

North star flow: goal intake → adaptive plan → day-of coaching

The Problem

Consistency breaks when the plan can’t flex

Training plans are usually built for ideal conditions. In reality, people miss days, change schedules, feel soreness, travel, or lose motivation. When plans don’t adapt, users either abandon the plan or overcorrect.

The challenge is to support progress without requiring perfect discipline.

Decision fatigue: “what should I do today?”

Drop-off moments: missed day → guilt → abandonment

Research Insights

People want coaching, not more tracking

Plans need “make-up” logic Missing a day shouldn’t reset the whole journey—users need a safe, judgment-free way back in that protects their progress without demanding perfection.

Recovery is the blind spot People struggle to understand their own soreness, sleep, and training load—so they either push too hard or stop entirely when things feel off.

Motivation is contextual Small wins, streak protection, and real-time guidance on what to do today matter more to consistency than long-term dashboards or abstract goal tracking.

Insight synthesis board

Behavior loop: cue → action → reward → identity

Defining the Opportunity

A flexible plan with one clear next step

The concept centers on reducing “what do I do today?” into a single adaptive recommendation that considers goal, time, recovery, and history.

Design principles

Keep the user in control: suggestions, not prescriptions
Be explicit about trade-offs: speed vs recovery vs consistency
Make re-entry easy: protect streaks and offer “return” workouts
Reward consistency, not perfection

Opportunity framing + job stories

Information architecture: Today, Plan, Insights, Gear

Design Exploration

An AI companion embedded in the Nike ecosystem

Goal intake A short onboarding flow translates intent into real constraints: available time, training frequency, equipment, injury context, and personal preferences.

Adaptive “Today” card One daily recommendation with a plain-language explanation and clear alternatives—shorter, easier, or more intense—so the decision is already made for you.

Recovery-aware adjustments When users report soreness or low energy, the system adapts training load and surfaces active recovery options to keep the plan on track.

Flow: setup → first week plan → day-of changes

Wireframes: Today card + alternatives + rationale

Test & Iterate

Prototype testing: Today card interaction and swap behavior

Iteration: adaptive plan logic and re-entry flow

Final Product

Clarity first: one next step, with context

The final concept emphasizes a lightweight daily flow: open the app, see one recommended next action, and start. Users can swap the workout, shorten it, or convert to recovery with transparent impact on the plan.

The companion also supports “make-up logic” after missed days by reshaping the plan rather than restarting it.

High-fidelity mocks: Today card + workout detail

Make-up logic: missed day → plan reshapes

Impact

Supporting consistency through adaptability

Intended outcomes: reduced decision fatigue, smoother re-entry after missed days, and clearer feedback loops that prioritize consistency.

Success metrics (concept): weekly active training + plan adherence

Qualitative outcomes: confidence + reduced guilt

Reflection

Designing AI for trust requires legibility

The key design challenge is not generating a plan—it’s communicating why a recommendation changed and what the user can do about it. This concept prioritizes transparency and user control to build trust in adaptive coaching.

What I’d test next: explanation UI + swap behavior

Future: cross-sport planning and community accountability