From Wireframes to Launch in 2 Weeks: Designing Video Registration

An early-stage hiring platform needed to register users through: recording a video, reviewing it, and then sign up with additional profile information. The team had wireframes, but they were fragmented. I had 2 weeks to make this feel intuitive, not overwhelming.

ROLE

UX/UI Designer (solo)

July to August 2024

2 weeks

2 weeks

2 weeks

SHIPPED

Launched August 2024

IMPACT

Redesigned the full post-registration system in 2025

Hero Image
Hero Image
Hero Image
Hero Image
Hero Image
Hero Image

The Problem

Fragmented Wireframes, No Guiding Hand

The existing wireframes had structure but lacked clarity

Users didn't know why they were recording a video

Users didn't know why they were recording a video

Users didn't know what to say

Users didn't know what to say

Users had no opportunity to review or redo their video

Users had no opportunity to review or redo their video

The flow jumped abruptly from recording to sign-up

The flow jumped abruptly from recording to sign-up

My Approach

Align First, Design Second

I spent week 1 aligning with the founder on three questions:

What does the platform actually need from this video?

What does the platform actually need from this video?

What would make a user feel confident recording?

What would make a user feel confident recording?

Where do users get stuck or abandoned?

Where do users get stuck or abandoned?

His answer was clear

CEO
We need people to feel like we're guiding them through this, not testing them

The Flow I Designed

Informative Recording to Signup

The Pitch

"Why do I need to record a video?"

I designed an informational pop-up that explained:

Why the platform needs this

What happens with the video

Why it matters

This step removes the "why am I doing this?" friction before it starts.

The Recording Interface

Is it recording? How much time do I have?

The interface showed:

Large camera preview (they can see themselves)

Clear timer, record/stop buttons

Buttons for video tops and example video

The Pitch

I designed an informational pop-up that explained:

Why the platform needs this

What happens with the video

Why it matters

This step removes the "why am I doing this?" friction before it starts.

The Recording Interface

The interface showed:

Large camera preview (they can see themselves)

Clear timer, record/stop buttons

Buttons for video tops and example video

Tested 3 control layouts

Shipped the mobile-overlay (Version C) because it matched users’ camera mental model and scaled cleanly to mobile.

Feature Image

Version C (Shipped)

All UI inside the video frame (familiar mobile pattern)

Feature Image

Version C (Shipped)

All UI inside the video frame (familiar mobile pattern)

Feature Image

Version C (Shipped)

All UI inside the video frame (familiar mobile pattern)

Feature Image

Version A (Hybrid)

Prompts inside, controls outside (split user focus)

Feature Image

Version A (Hybrid)

Prompts inside, controls outside (split user focus)

Feature Image

Version A (Hybrid)

Prompts inside, controls outside (split user focus)

Feature Image

Version B (External)

All UI outside and controls outside for users to focus on recording

Feature Image

Version B (External)

All UI outside and controls outside for users to focus on recording

Feature Image

Version B (External)

All UI outside and controls outside for users to focus on recording

Review & Redo

After recording, users could:

Watch their full video

Redo it as many times as they wanted (no limit)

Feel confident before moving forward

This step is critical because it shifts power to the user. We are not forcing them to submit something they're uncomfortable with.

Sign-Up + Profile Setup

After they approved their video, they filled out:

Basic info (name, location, role interest)

A few profile questions

These questions were short and scannable. The goal was capturing just enough to build their profile, not exhausting them.

Review & Redo

What if I don't like it? Can I try again?

After recording, users could:

Watch their full video

Redo it as many times as they wanted (no limit)

Feel confident before moving forward

This step is critical because it shifts power to the user. We are not forcing them to submit something they're uncomfortable with.

Sign-Up + Profile Setup

Now what? What else do they need from me?

After they approved their video, they filled out:

Basic info (name, location, role interest)

A few profile questions

These questions were short and scannable. The goal was capturing just enough to build their profile, not exhausting them.

The Constraint

Speed Over Research

I had 2 weeks to ship, which meant no time for user testing. To mitigate risk without data, I:

Aligned on Empathy: Asked the founder "Where do you see users get stuck?" to identify anxiety points.

Aligned on Empathy: Asked the founder "Where do you see users get stuck?" to identify anxiety points.

Relied on conventions: Used familiar patterns (like the mobile-style recording UI) to reduce the need for learning.

Relied on conventions: Used familiar patterns (like the mobile-style recording UI) to reduce the need for learning.

Built for Iteration: Designed the flow knowing we would refine it after launch based on real user behavior.

Built for Iteration: Designed the flow knowing we would refine it after launch based on real user behavior.

The video registration launched and drop-off happened. The founder needed to know if users would complete video registration upfront. A year later they hired me back to redesign the post-registration experience around that learning

What I Learned

Shipping Beats Perfection

[01]

Pharmacy side

Add onboarding guidance

Product templates for common medicines

Real analytics dashboard focused on operator questions

[03]

Integration

SMS tied to order lifecycle (not just a tool)

Multi-location support for larger pharmacy chains

[02]

Consumer side

Run research with both operators and patients

Test which discovery method converts better: pharmacies-first vs. category-first

Add trust signals: pharmacy verification, delivery times, ratings

Build the prescription ordering flow (I only hinted at it)

[01]

Pharmacy side

Add onboarding guidance

Product templates for common medicines

Real analytics dashboard focused on operator questions

[03]

Integration

SMS tied to order lifecycle (not just a tool)

Multi-location support for larger pharmacy chains

[02]

Consumer side

Run research with both operators and patients

Test which discovery method converts better: pharmacies-first vs. category-first

Add trust signals: pharmacy verification, delivery times, ratings

Build the prescription ordering flow (I only hinted at it)

[01]

Confidence > Features

User anxiety is the biggest blocker in complex flows

Solving for fear (with guidance and redos) mattered more than adding more recording features.

[02]

Alignment Prevents Rework

Spending Week 1 aligning on the founder’s goals meant zero major revisions later.

We shipped in 2 weeks because we agreed on "good enough" early.

[03]

Ship to Learn

Shipping something thoughtful teaches you what to build next.

A year later, the company used those learnings to pivot and hired me back to redesign the post-registration experience based on this foundation.

[01]

Confidence > Features

User anxiety is the biggest blocker in complex flows

Solving for fear (with guidance and redos) mattered more than adding more recording features.

[02]

Alignment Prevents Rework

Spending Week 1 aligning on the founder’s goals meant zero major revisions later.

We shipped in 2 weeks because we agreed on "good enough" early.

[03]

Ship to Learn

Shipping something thoughtful teaches you what to build next.

A year later, the company used those learnings to pivot and hired me back to redesign the post-registration experience based on this foundation.