Outfitsy is a fashion app that helps people create everyday outfits from a digital wardrobe. Its swipe-based builder makes it easier to explore combinations, plan what to wear and move beyond the daily “nothing to wear” dilemma.

Outfitsy is a fashion app that helps people create everyday outfits from a digital wardrobe. Its swipe-based builder makes it easier to explore combinations, plan what to wear and move beyond the daily “nothing to wear” dilemma.

Team

CEO, Developers, ASO Manager

CEO, Developers, ASO Manager

Timeline

Dec 2024 – Aug 2026

Dec 2024 – Aug 2026

Role

Sole Product Designer

Sole Product Designer

01. Project Overview

From first concept to 150K+ users worldwide for the six month

I was the sole Product Designer and led product decisions from concept and MVP definition through launch and post-release development. I worked directly with the CEO and developers, and before release with an ASO manager, preparing the App Store screenshots from a brief.

I was the sole Product Designer and led product decisions from concept and MVP definition through launch and post-release development. I worked directly with the CEO and developers, and before release with an ASO manager, preparing the App Store screenshots from a brief.

02. Challenge

Planning outfits takes time, and wardrobe apps often ask for too much setup first.

People had no habit of planning outfits, no easy way to keep track of their wardrobe, and no tool to see a combination without handling the clothes. The challenge was to make outfit creation effortless without becoming another wardrobe-management tool, keeping the core interaction simple while leaving room for planning, personalization and wardrobe organization.

People had no habit of planning outfits, no easy way to keep track of their wardrobe, and no tool to see a combination without handling the clothes. The challenge was to make outfit creation effortless without becoming another wardrobe-management tool, keeping the core interaction simple while leaving room for planning, personalization and wardrobe organization.

03. Approach

Define a minimal product, test it before release, measure everything after.
  • MVP. A base of standard items, an outfit builder, ready-made outfits, an in-app camera, a week and month planner and feedback forms, with no subscription or paid features. It launched about six months after the project began.

  • Group testing. About 12 people with different phones, professions and ages tested the MVP offline in one room. It produced 22 findings, each with a proposed solution, priority and estimates for iOS and Android.


  • Analytics. Five areas: basic usage, feature interaction (onboarding, outfit creation, planning, photo processing quality), retention, user feedback and technical performance.

  • MVP. A base of standard items, an outfit builder, ready-made outfits, an in-app camera, a week and month planner and feedback forms, with no subscription or paid features. It launched about six months after the project began.

  • Group testing. About 12 people with different phones, professions and ages tested the MVP offline in one room. It produced 22 findings, each with a proposed solution, priority and estimates for iOS and Android.


  • Analytics. Five areas: basic usage, feature interaction (onboarding, outfit creation, planning, photo processing quality), retention, user feedback and technical performance.

04. Design Decisions

Outfits first, wardrobe second

Solution: The swipe-based builder is the core interaction, and a curated starter wardrobe lets users create outfits before uploading their own clothes.

Evidence: Testers wanted to explore before investing time, so onboarding became simpler and more useful, and the wardrobe could be filled right away.

Solution: The swipe-based builder is the core interaction, and a curated starter wardrobe lets users create outfits before uploading their own clothes.

Evidence: Testers wanted to explore before investing time, so onboarding became simpler and more useful, and the wardrobe could be filled right away.

Fast creation with hands-on control

Solution: Accessories can be added, moved and resized on the outfit collage, and one click picks an outfit at random.

Evidence: Swipe worked well for combinations, but testers wanted more control over accessories. Random outfits encouraged more interaction with the builder.

Solution: Accessories can be added, moved and resized on the outfit collage, and one click picks an outfit at random.

Evidence: Swipe worked well for combinations, but testers wanted more control over accessories. Random outfits encouraged more interaction with the builder.

Adding real clothes with less effort

Solution: Photos from the camera or gallery go through AI background removal and image processing, so any piece becomes a clean item for collages.

Evidence: In testing, people photographed clothes quickly and imperfectly. We tracked how often the same photo was sent for processing to measure recognition quality.

Solution: Photos from the camera or gallery go through AI background removal and image processing, so any piece becomes a clean item for collages.

Evidence: In testing, people photographed clothes quickly and imperfectly. We tracked how often the same photo was sent for processing to measure recognition quality.

Improving after launch

Solution: A restructured wardrobe with category filters, several outfits per day, sharing to social networks and messengers, and real photos on outfit collages.

Evidence: Testers asked for more than one outfit a day and for sharing. The rest came from analytics and user feedback.

Solution: A restructured wardrobe with category filters, several outfits per day, sharing to social networks and messengers, and real photos on outfit collages.

Evidence: Testers asked for more than one outfit a day and for sharing. The rest came from analytics and user feedback.

Core value first, complexity only when it earns

We chose not to build in-app communities, to avoid overloading users, and not to make manual collage the main way to build outfits, because it slowed assembly. Complex multi-layered outfits were postponed: they needed a deep rework of the swipe builder and added complexity for users who did not need them, so we treated them as a hypothesis to study first.

We chose not to build in-app communities, to avoid overloading users, and not to make manual collage the main way to build outfits, because it slowed assembly. Complex multi-layered outfits were postponed: they needed a deep rework of the swipe builder and added complexity for users who did not need them, so we treated them as a hypothesis to study first.

Mockup image

05. Results

The product launched approximately six months after the project began, followed by user testing and further iterations ahead of the public iOS and Android release. Rather than adding complexity for the sake of more features, I focused on balancing speed, flexibility and usability as the product grew.

  • Audience: iOS users were mostly women aged 20–35, while Android users were mostly teenagers.

  • Onboarding: the starter wardrobe and simpler onboarding helped new users get started, according to onboarding analytics and feedback.

  • Engagement: one-click random outfits encouraged more interaction with the builder.

The product launched approximately six months after the project began, followed by user testing and further iterations ahead of the public iOS and Android release. Rather than adding complexity for the sake of more features, I focused on balancing speed, flexibility and usability as the product grew.

  • Audience: iOS users were mostly women aged 20–35, while Android users were mostly teenagers.

  • Onboarding: the starter wardrobe and simpler onboarding helped new users get started, according to onboarding analytics and feedback.

  • Engagement: one-click random outfits encouraged more interaction with the builder.

0

users in the first week

0

users worldwide within the first six months