Digital retention strategy for the fragrance subscription

I worked at Scentbird for two years and combined product manager and designer roles.

Scentbird is a New-York-based fragrance subscription with over 300,000 active subscribers.

Scentbird physical product

Physical product heavily affects retention

User problem

Business goal

Help users discover the world of fragrance.

Key metrics

Retention, LTV


My team could affect only digital UX/UI, while shipping, CS, and the physical product itself play a major role in user retention.

Projects scoring

Monthly billing cycles meant longer testing times

Long A/B testing cycles (sometimes months) to gain statistically significant results due to a relatively small number of user interactions.

Product designer/manager

I managed a product team of front-end and back-end developers, designer, data scientist, analyst, QA.

I did all the research myself: user interviews, data analysis, experiment design, remote user testing, etc.

I also worked on key concepts for future designs, built prototypes, and later collaborated with UI designers and team engineers to develop prototypes into a final product.

Supervised and helped designers from other team to build the UI library in Figma. We occasionally collaborated with the Humble Team (agency).

Hypotheses source

Most of the research I did at the time was about either reason to unsubscribe or why power users are staying with Scentbird for longer and if I can find patterns in their behavior that can be introduced to fast-churning users. Things I’ve learned from users gave me lots of hypotheses to test: navigation, product card redesign, homepage redesign, new onboarding, new recommendations algorithm.

Design process

Qualitative data

Gather insights (phone interviews, on-site interviews, remote user testing, card sorting, etc.)

Quantitative research

Reqired to support ideas (database queries, information from Customer support, data from analysts)


Layouts or direct cooperation with front-end developers if possible

Preliminary testing

If applicable

A/B testing

Data analysis, follow-ups

How to pick the perfect fragrance online

Scentbird: building blocks of the perfect fragrance recommender

Fragrance recommendations based on the real-life experience


Picking the right fragrance online is tough, but the result is crucial both for customers and service. To build a proper recommender, we created data pipelines, gathered more data to train a new model, and completely redesigned the product card.

Dive into the details
Scentbird: Mobile navigation redesign

Information architecture improvements: 20% faster task completion rate


We removed 50% of UI elements from mobile navigation without dropping any metrics and made it easier to understand for 80% of users.

Mobile navigation redesign

Homepage redesign + onboarding: focus on a fragrance


The new personalized modular system helped registered users to get the most of their subscription while increasing LTV.

Compare before and after

Nobody does it alone

Most of the UI was created in tight collaboration with Karine Arutyunyan, Anna Voshkarina and the Retention Team at Scentbird.

The new visual style was created in collaboration with my friends from the Humble Team studio.

A tremendous amount of work on icons and brand visuals was produced by the internal creative team at Scentbird.