AS

AS

Diagnosing and mending the in-store personalization gap

Diagnosing and mending the in-store personalization gap

Helping retailers shift from extraction to exchange with AI

Helping retailers shift from extraction to exchange with AI

This capstone project introduces Cerulean, an AI-enabled personalization tool that addresses the 91% of apparel shoppers who feel overwhelmed in-store.

Design Lead

|

8 months

|

Designed for ENAiBLE, Carnegie Mellon University’s retail collective

As Design Lead, I drove our rapid prototyping strategy, guiding 20+ prototypes to test shopper needs and refine core features. I defined Cerulean’s visual design language for cohesion across flows and touchpoints, and advised on foundational research strategy to align insights with design direction.

↓ Scroll down to read the Cerulean story ↓

This capstone project introduces Cerulean, a personalization tool that addresses the 91% of apparel shoppers who feel overwhelmed in-store. Cerulean helps shoppers stay in control while giving retailers clearer ways to turn shopper data into value.

Design Lead

|

8 months

|

Designed for ENAiBLE, Carnegie Mellon University’s retail collective

As Design Lead, I drove our rapid prototyping strategy, guiding 30+ prototypes to test shopper needs and refine core features. I defined Cerulean’s visual design language for cohesion across flows and touchpoints, and advised on foundational research strategy to align insights with design direction and business goals.

Retail is more advanced than ever, yet most shoppers describe in-store shopping negatively..

...despite wanting similar things, shoppers and retailers seem to be at odds...

“I just want to find the right product for ME, within budget which will look and feel good in.”

“I want to help shoppers find the right product from my brand, so that I can become their go-to.”

This is the story of how I led the design and strategy of Cerulean, a system built to bridge the apparel retailer-shopper disconnect in 2025.

Retail is more advanced than ever, yet most shoppers describe in-store shopping negatively..

...despite wanting similar things, shoppers and retailers seem to be at odds...

“I just want to find the right product for ME, within budget which will look and feel good in.”

“I want to help shoppers find the right product from my brand, so that I can become their go-to.”

This is the story of how I designed a system built to bridge the apparel retailer-shopper disconnect in 2025.

01-Context: Apparel retail has an in-store personalization problem.

01-Context: Apparel retail has an in-store personalization problem.

Online, we’re used to personalization that matches our exact taste with optimal timing.

Online, we’re used to personalization that matches our exact taste with optimal timing.

In-store though? You’re kind of on your own.

In-store though? You’re kind of on your own.

Spotify and Netflix? Top tier. Uniqlo digital? Solid. Uniqlo in-store? Godspeed.

Spotify and Netflix? Top tier. Uniqlo digital? Solid. Uniqlo in-store? Godspeed.

02-Our Diagnosis: The retailer-shopper relationship needs to be reconstructed.

02-Our Diagnosis: The retailer-shopper relationship needs to be reconstructed.

Our client suspected that retailers were stalled by two constraints: organizational silos and legacy tech. We uncovered something deeper: a foundational misalignment in how retailers are personalizing and what shoppers actually need.

Our client suspected that retailers were stalled by two constraints: organizational silos and legacy tech. We uncovered something deeper: a foundational misalignment in how retailers are personalizing and what shoppers actually need.

You see, despite internal challenges, retailers are actively pursuing personalization. Most through data-enabled strategies like clienteling and targeted communications. But neither is resonating with shoppers, who we found just want experiences driven by Autonomy, Confidence, and Respect (which we modeled as ACR).

You see, despite internal challenges, retailers are actively pursuing personalization. Most through data-enabled strategies like clienteling and targeted communications. But neither is resonating with shoppers, who we found just want experiences driven by Autonomy, Confidence, and Respect (which we modeled as ACR).

I, shopper, want shopping enabled by…

I, shopper, want shopping enabled by…

AUTONOMY

AUTONOMY

access to information about a product and control over how
I access it

access to information about a product and control over how
I access it

clienteling

clienteling

CONFIDENCE

CONFIDENCE

trust in the product's fit, function, price

trust in the product's fit, function, price

clienteling

clienteling

RESPECT

RESPECT

clarity in how my data is used and meaningful value in return

clarity in how my data is used and meaningful value in return

targeted comms

targeted comms

Clienteling and targeted comms undermine all three needs. Clienteling is perceived as a gamble and targeted comms as risky rather than valuable.

Clienteling and targeted comms undermine all three needs. Clienteling is perceived as a gamble and targeted comms as risky rather than valuable.

How did retailers miss the mark? Well, they got a bit tunnel visioned. As it turned out, retailers are quite data poor in-store.

How did retailers miss the mark? Well, they got a bit tunnel visioned. As it turned out, retailers are quite data poor in-store.

Retailers are reliant on shoppers sharing their loyalty accounts with each trip, for example.

Retailers are reliant on shoppers sharing their loyalty accounts with each trip, for example.

To compensate, retailers are gathering as much shopper data as they can. However, truly complete data is essentially impossible to achieve.

To compensate, retailers are gathering as much shopper data as they can. However, truly complete data is essentially impossible to achieve.

This is because shopper data is fragmented. People don’t just shop with one brand - they hop across brands trying to find their perfect product.

This is because shopper data is fragmented. People don’t just shop with one brand - they hop across brands trying to find their perfect product.

In the process of chasing data, retailers are ignoring shopper’s need for ACR and alienating them.

In the process of chasing data, retailers are ignoring shopper’s need for ACR and alienating them.

If retailers want the missing data, they need to shift their approach from extraction to value. Because when data is the product, the only way to earn it is by designing for the needs of the user.

If retailers want the missing data, they need to shift their approach from extraction to value. Because when data is the product, the only way to earn it is by designing for the needs of the user.

We realized, no single retailer can easily fix this. However, a neutral, shopper-centered bridge just might.

We realized, no single retailer can easily fix this. However, a neutral, shopper-centered bridge just might.

We imagined a bridge capable of aggregating shopper data for retailers and providing truly valuable personalization for shoppers.

We imagined a bridge capable of aggregating shopper data for retailers and providing truly valuable personalization for shoppers.

03-Prototyping: Designing personalization rooted in shopper needs.

03-Prototyping: Designing personalization rooted in shopper needs.

To get there, we set out to map the line between helpful and invasive personalization.

To get there, we set out to map the line between helpful and invasive personalization.

We focused on three key areas: Autonomy, Confidence, and Respect – each shaped by smaller boundaries shoppers navigate: control, choice, comfort, information, identity, data requests, and the “creepiness” factor.

We focused on three key areas: Autonomy, Confidence, and Respect – each shaped by smaller boundaries shoppers navigate: control, choice, comfort, information, identity, data requests, and the “creepiness” factor.

To explore each, we ran over 25 tests.

To explore each, we ran over 25 tests.

Autonomy

How much control do shoppers really want?

Autonomy

How much control do shoppers really want?

Autonomy

What if we gave shoppers more choice, not less?

Autonomy

What if we gave shoppers more choice, not less?

Autonomy

What type of technologies are shoppers actually willing to use?

Autonomy

What type of technologies are shoppers actually willing to use?

Confidence

How far do shoppers want to stretch their style?

Confidence

How far do shoppers want to stretch their style?

Confidence

How much information is too much?

Confidence

How much information is too much?

Respect

What solutions make shoppers uncomfortable?

Respect

What solutions make shoppers uncomfortable?

How much control do shoppers really want?

How much control do shoppers really want?

Respect

How much is too much to ask of shoppers?

Respect

How much is too much to ask of shoppers?

The essential quesstions behind our seven most influential tests.

The essential quesstions behind our seven most influential tests.

From our insights, we built six design principles and refined them into four concepts. The concepts that simplified prep, removed in-store friction, and empowered shoppers became the core of Cerulean.

From our insights, we built six design principles and refined them into four concepts. The concepts that simplified prep, removed in-store friction, and empowered shoppers became the core of Cerulean.

We prioritized and finalized features by analyzing feasability and desirability. The result: A smarter prep tool for in-store shopping

We prioritized and finalized features by analyzing feasability and desirability. The result: A smarter prep tool for in-store shopping

04-Our Solution: Cerulean in Action.

04-Our Solution: Cerulean in Action.

Let me introduce Cerulean: a mobile app designed to make shopping intentional and personal.

Let me introduce Cerulean: a mobile app designed to make shopping intentional and personal.

But first, let's look at the pain points plaguing the current in-store shopping journey. Each of these moments are a result of synthesis of shopper experiences.

But first, let's look at the pain points plaguing the current in-store shopping journey. Each of these moments are a result of synthesis of shopper experiences.

Pain point #1

Overwhelming results in planning

Too many irrelevant or out-of-stock items.

Pain point #2

Hard to translate ideas into results

Shoppers struggle to turn a mental image or style inspiration into the right products.

Pain point #3

Out-of-stock surprises

Trips derailed by unavailable items.

Pain point #4

Low product discovery

Relevant items are hidden or missed in-store.

Pain point #5:

Shopping fatigue

Time wasted, energy drained.

Pain point #1

Overwhelming results in planning

Too many irrelevant or out-of-stock items.

Pain point #2

Hard to translate ideas into results

Shoppers struggle to turn a mental image or style inspiration into the right products.

Pain point #3

Out-of-stock surprises

Trips derailed by unavailable items.

Pain point #4

Low product discovery

Relevant items are hidden or missed in-store.

Pain point #5:

Shopping fatigue

Time wasted, energy drained.

Each pain point became a design opportunity. Here’s how those opportunities turned into Cerulean’s key features.

Each pain point became a design opportunity. Here’s how those opportunities turned into Cerulean’s key features.

Contextual Search

What it is: Search that considers what shoppers need, when and where they’ll wear it, and under what conditions to surface accurate, in-stock options across brands.

Solving for (Pain Point #1): Cuts through overwhelming results in planning by filtering out irrelevant and out-of-stock items.

Search Refinement

What it is: An intuitive, Pinterest-like visual process that lets shoppers narrow search results based on the details that matter most to them.

Solving for (Pain Point #2): Bridges the gap between style ideas and tangible results with an intuitive, visual narrowing process.

Curated Match List

What it is: Prioritized, location-aware product list showing in-stock items that meet a budget, size, and context, with expandable product cards.

Solving for (Pain Point #3): Eliminates out-of-stock surprises by giving shoppers a ready-to-buy, prioritized list.

Smart Filters

What it is: Smart filters narrow results by size, style, budget, and context, remembering preferences so shoppers don’t have to.

Solving for (Pain Point 4): Reduces mental effort by remembering preferences and updating automatically, keeping results relevant without repetition.

Plan Map

What it is: The map view plots matched items across nearby stores with real-time availability.

Solving for (Pain Point #5): Reduces shopping fatigue by helping plan efficient trips and avoid wasted stops.

Contextual Search

What it is: Search that considers what shoppers need, when and where they’ll wear it, and under what conditions to surface accurate, in-stock options across brands.

Solving for (Pain Point #1): Cuts through overwhelming results in planning by filtering out irrelevant and out-of-stock items.

Search Refinement

What it is: An intuitive, Pinterest-like visual process that lets shoppers narrow search results based on the details that matter most to them.

Solving for (Pain Point #2): Bridges the gap between style ideas and tangible results with an intuitive, visual narrowing process.

Curated Match List

What it is: Prioritized, location-aware product list showing in-stock items that meet a budget, size, and context, with expandable product cards.

Solving for (Pain Point #3): Eliminates out-of-stock surprises by giving shoppers a ready-to-buy, prioritized list.

Smart Filters

What it is: Smart filters narrow results by size, style, budget, and context, remembering preferences so shoppers don’t have to.

Solving for (Pain Point 4): Reduces mental effort by remembering preferences and updating automatically, keeping results relevant without repetition.

Plan Map

What it is: The map view plots matched items across nearby stores with real-time availability.

Solving for (Pain Point #5): Reduces shopping fatigue by helping plan efficient trips and avoid wasted stops.

05-Impact: Cerulean is designed to benefit both shoppers and retailers.

05-Impact: Cerulean is designed to benefit both shoppers and retailers.

From day one, we designed Cerulean to work for both sides of the shopping experience. Retailers were part of the process - sharing where they saw opportunity, and helping shape a tool that delivers measurable impact.

From day one, we designed Cerulean to work for both sides of the shopping experience. Retailers were part of the process - sharing where they saw opportunity, and helping shape a tool that delivers measurable impact.

An app that delivers personalized style recommendations while also aggregating prices, store locations, and retailer availability could be a game-changer - especially for reviving foot traffic in physical retail spaces.

– Former Senior Manager at Macy's, Inc.

An app that delivers personalized style recommendations while also aggregating prices, store locations, and retailer availability could be a game-changer - especially for reviving foot traffic in physical retail spaces.

– Former Senior Manager at Macy's, Inc.

Retailer value at a glance:

Retailer value at a glance:

06-Personal Growth: How Cerulean shaped me as a leader and designer.

06-Personal Growth: How Cerulean shaped me as a leader and designer.

As the most experienced member of the team, I had to lead with intention while staying flexible. I learned to adapt to surprises, support my teammates, and keep our work grounded in the real needs of shoppers and retailers.

As the most experienced member of the team, I had to lead with intention while staying flexible. I learned to adapt to surprises, support my teammates, and keep our work grounded in the real needs of shoppers and retailers.

How I learned to lead

  • Balancing providing direction with space for collaboration, even when it took extra time

  • Giving my team feedback on things I know well, like research, but also trust them to lead in their own areas

  • Advocating for my team so their work and ideas were seen

How I learned to lead

  • Balancing providing direction with space for collaboration, even when it took extra time

  • Giving my team feedback on things I know well, like research, but also trust them to lead in their own areas

  • Advocating for my team so their work and ideas were seen

How I learned to work:

  • Run quick, scrappy tests without stressing over one big, formal protocol

  • Let early results point me in the right direction, even if they weren’t complete

  • Ask for critique, adapted to it quickly, and use it to make the work stronger

How I learned to work:

  • Run quick, scrappy tests without stressing over one big, formal protocol

  • Let early results point me in the right direction, even if they weren’t complete

  • Ask for critique, adapted to it quickly, and use it to make the work stronger

Ashley Sánchez

Thanks for stopping by!

Ashley Sánchez

Thanks for stopping by!

Ashley Sánchez

Thanks for stopping by!