A+Link

Contributions

The original A+Link MVP had real functionality — but no design coherence. Students couldn't find tutors easily, didn't trust what they found, and abandoned the booking flow before completing it. Trust was the core problem, and the design was actively working against it.

• Booking confusion: The booking flow had too many steps with no clear progress indicator — students dropped off before confirming.

• No trust signals: Tutor profiles had no credentials, ratings, or social proof — students felt they were booking blind.

• Complex navigation: Student and tutor flows were entangled in a maze — switching between modes was disorienting.

• Tutor discovery friction: No filtering, no smart matching — students scrolled through an undifferentiated list.

• Missing review system: No way to see feedback from previous students — removing the one signal that builds platform trust.

I interviewed students and tutors to understand where the original app broke down — from confusing booking flows to the absence of trust signals that made students hesitant to commit to any tutor.

Competitive Analysis

Analysis of Wyzant, Tutor.com, Preply, and Chegg Tutors revealed what patterns users already expect — and where there was whitespace to differentiate through trust and simplicity.

Personas

A+Link is a two-sided marketplace — I designed for both the student seeking help and the tutor building their practice. Their needs are distinct but the platform must serve both without friction.

Houda, 21

University Student · Struggling in Calculus

"I just need someone who actually explains things, not just solves the problems for me."

Imane, 34

Graduate Tutor · Building Clientele

"I want students to find me easily — and I want the platform to look as professional as I am."

Information Architecture

The original app's navigation was a maze. I rebuilt the IA from scratch — separating the student and tutor experiences while keeping core flows intuitive for both user types.

Before & After

The original MVP had real functionality but no design coherence. Every decision in the redesign was driven by a specific research finding or usability failure.

Before — Original MVP

After — Redesigned

Design System

A complete design system replacing the ad-hoc styling of the original — consistent typography, color, components, and states that make every interaction feel intentional.

High-Fidelity Design

The redesigned screens solve specific problems identified in research — clear tutor cards with trust signals, a streamlined booking flow, and a student dashboard that shows progress at a glance.

• Complete design system delivered — replaced all ad-hoc MVP styling with a consistent, scalable component library.

• Trust signals introduced across every tutor-facing screen — credentials, ratings, and social proof now visible before booking.

• Database schema reviewed with Claude — caught relational issues and edge cases before they reached production.

• IA rebuilt from scratch — student and tutor flows now separated cleanly, with shared infrastructure underneath.

A+Link taught me that trust is a design decision, not a feature. Every screen either builds confidence or erodes it. The redesign wasn't about making things look better — it was about making users feel safe enough to act. That shift in framing changed every decision, from component hierarchy to the order of fields in the booking flow.

• Trust is the product: On two-sided platforms, the experience is only as strong as the weakest trust signal. Designing for safety unlocked engagement.

• Design and dev must talk: Using Claude to review the schema while designing surfaced conflicts early — before a single line of production code was written.

• Two personas, one IA: Designing for two fundamentally different user types on the same platform forced architectural decisions that a single-user project never would.

• Before/after discipline: Mapping every redesign decision back to a specific research failure kept scope tight and every change defensible.