thredUP Personalization

About thredUP:

thredUP is Goldman-Sachs backed fashion resale website for consumers to buy and sell secondhand clothing online. At any given time, the site can have 1 million unique items for sale.



Product Manager, Design Leadership, Copywriting


We wanted to improve the user-reported feelings of “being overwhelmed” by the amount of items to sort though and improve add-to-cart rates on the website. Our hypothesis was that personalization would improve both of these customer experiences.


  • Fully responsive personalization hub space that includes new features:

    • saved sizes

    • saved brands

    • favoriting

    • “closet” - previous purchase history plus styling recommendations to shop

    • computer vision driven AI recommendations based on shopping history

  • Hub page ranked #6 of all page views on the site

  • Favoriting responsible for 2% lift in conversion

  • Saved sizes responsible for 5% lift in conversion

mythredupTODAY (1).gif


The entire leadership team of thredUP had ideas about how to personalize thredUP. I pared down the feature list by 60% by gathering user feedback (MaxDiff survey, n=1600).

Partnered with third party for AI-driven recommendations (Mad Street Den).

I oversaw a team of designers and researchers throughout all stages of the design, and wireframed the earliest IA.

my thredup early wires.png

Testing 1-2-3

We brought users onsite during the prototyping and design phase. We iterated based on that feedback in order to reach final designs. Once we passed those off, we dogfooded the features all internally before releasing to a small subset of users. 20% of this subset received a pop-up survey asking for feedback on the recommendations.

thredup recommendation.jpeg
thredup recommendations feedback.png
feedback from auryc.png


  • my thredUP was released to 100% of users in February of 2017 with positive reception. It has since (as of Oct 2018) been re-branded to Dressing Room.

Nothin’ can stop me,

I’m all the way up!