Idesignproductsthatmovepeople—literally.

User interviews — 8-12 users per project, split by persona (first-time users vs. power users, patients vs. clinicians, merchants vs. customers). I ask behavioral questions ('Walk me through the last time you...') not hypotheticals ('Would you use...?').

• Competitive analysis — I study 3-5 competitors to identify industry patterns (what users expect) and gaps (opportunities to differentiate).
• Support ticket analysis — High support volume = design failure. I analyze tickets by category (product confusion, technical errors, process questions) to identify what needs fixing vs. what needs better communication.
• Stakeholder alignment — I work with product managers, engineers, and business teams to understand: What metrics matter? What constraints exist (technical, regulatory, budget)? What's the real problem we're solving?

Define the problem clearly — I write a one-sentence problem statement that everyone can rally behind. Example: 'Merchants lose 60% of sales inquiries because they can't respond fast enough on WhatsApp.'

• Map user journeys — I document the current state (where users struggle) vs. ideal state (what success looks like). This reveals friction points and opportunities.
• Prioritize features ruthlessly — I use the RICE framework (Reach × Impact × Confidence / Effort) to rank features. High reach + high impact + low effort = build first. Low confidence or high effort = validate with prototype before committing.
• Set success metrics — I define what 'working' means before we build. Not vanity metrics (screen views), but outcomes (conversion rate, time to complete task, support ticket reduction, NPS).

Start low-fidelity — Wireframes first. I validate structure and flow before investing in visual polish. If wireframe doesn't solve the problem, high-fidelity won't save it.

• Design systems thinking — I build reusable components and patterns from day one. Every button, form field, and card follows a system. This speeds up future work and ensures consistency.
• Motion design as a deliverable — Transitions, micro-interactions, and loading states aren't afterthoughts. I design them in After Effects or Framer Motion and ship them as code (Lottie files, Framer Motion components). Motion communicates state, guides attention, and builds trust.
• Accessibility by default — WCAG AA compliance, screen reader support, keyboard navigation, color contrast, mobile-first responsive design. I design for everyone, not just ideal users.
• Prototype with code — I use Claude Code, Windsurf, and Framer to build working prototypes — not clickable PDFs. This lets me test real interactions (animations, form validation, API calls) before engineering builds.

Usability testing — 5-8 users, task-based scenarios ('Upload a video and schedule it for tomorrow'). I watch where they struggle, what they misunderstand, what delights them. I'm looking for: Can they complete the task? How long does it take? Where do they get stuck?

• A/B testing — When there's uncertainty (e.g., 'Should onboarding show a tutorial or let users explore?'), I design both versions. Let data decide, not opinions.
• Prototype testing — Before engineering builds, I test interactive prototypes with real users. This catches UX issues early when they're cheap to fix.
• Heuristic evaluation — I audit designs against usability principles (Nielsen's 10 heuristics, Fitts's Law, progressive disclosure, etc.). Catches issues users might not articulate.

Annotated specs — Engineers get designs with notes: interaction states (hover, active, disabled), edge cases (what happens when there's no data, when API fails), design rationale (why this choice matters). This reduces back-and-forth and ensures quality.

• Design QA — I review builds before they ship. I check: Does it match the design? Are interactions smooth? Do edge cases work? Are there accessibility issues?
• Post-launch monitoring — I use Hotjar (heatmaps, session recordings), GA4 (funnels, drop-off points), and Mixpanel (cohort analysis) to track how users actually behave. I look for: Where do they drop off? What do they click that doesn't work? What's taking too long?
• Iterate based on data — Weekly reviews (quick wins), monthly deep dives (bigger changes). I don't design and forget — I design, ship, measure, iterate.