Depop, the London-based online fashion marketplace with 35 million registered users, has launched an innovative AI-powered description generator to streamline the selling process on its platform. The feature, unveiled in September as part of the company’s “CXO AI Playbook” initiative, uses image recognition and generative AI to automatically create product descriptions when sellers upload photos of items they want to sell.
According to Rafe Colburn, Depop’s Chief Product and Technology Officer, the company’s mission is to encourage consumers to “participate in the circular economy rather than buying new.” However, the time and effort required to list items for sale has been a significant barrier to platform adoption. The new AI tool addresses this challenge by reducing the manual work involved in creating listings.
The technology was developed in-house by Depop’s data science team, which trained large language models specifically for this purpose. In a strategic move in 2022, Depop relocated its data science team from the engineering group to the product side of the business, enabling faster feature releases and more agile development.
The AI description generator works seamlessly: sellers simply upload an image and click a “generate description” button. The system then uses image recognition and generative AI to create a complete product description and automatically populate item-attribute fields including category, subcategory, color, and brand. The technology incorporates relevant hashtags and colloquial language tailored to Depop’s user base. Colburn emphasized that “we’ve done a lot of prompt engineering and fine-tuning to make sure that the tone and style of the descriptions that are generated really fit the norms of Depop.”
Sellers retain full control, able to use the generated description as-is or modify it to their preferences. Even with modifications, the tool saves considerable time compared to starting from scratch. With approximately 180,000 new listings created daily on the platform, the impact is substantial. Since the September launch, Depop has observed “a real uplift in listings created, listing time, and completeness of listings,” though specific metrics weren’t yet available at the time of reporting.
The company, which was acquired by Etsy in 2021, continues to explore additional AI applications. Future initiatives include using AI to enhance photo quality and help buyers determine clothing fit more accurately, addressing two other significant pain points in the online secondhand fashion marketplace.
Key Quotes
By reducing that effort, we can make resale more accessible to busy people.
Rafe Colburn, Depop’s Chief Product and Technology Officer, explained the core motivation behind implementing AI tools. This quote underscores how AI is being used to democratize access to sustainable fashion by removing time barriers that prevent people from participating in the secondhand economy.
What we’ve tried to do is make it so that once people have photographed and uploaded their items, very little effort is required to complete their listing.
Colburn described the user experience goal for the AI description generator. This statement highlights the company’s focus on minimizing friction in the selling process, particularly for new sellers unfamiliar with the platform’s conventions.
We’ve done a lot of prompt engineering and fine-tuning to make sure that the tone and style of the descriptions that are generated really fit the norms of Depop.
Colburn revealed the technical approach behind the AI tool’s development. This quote is significant because it demonstrates that successful generative AI implementation requires more than just deploying off-the-shelf models—it demands careful customization to match specific community standards and expectations.
Aside from the direct user benefits in terms of efficiency and listing quality, we have also really demonstrated to ourselves that users value features that use generative AI to reduce effort on their end.
Colburn reflected on the broader lessons learned from the AI tool’s launch. This insight validates the company’s AI strategy and suggests that user acceptance of generative AI features is strong when they deliver tangible time savings and practical benefits.
Our Take
Depop’s AI implementation offers valuable lessons for the broader tech industry. The company’s approach—focusing on specific pain points rather than implementing AI for its own sake—demonstrates mature product thinking. By targeting the time-consuming task of writing product descriptions, Depop addresses a genuine barrier to platform participation.
What’s particularly impressive is the organizational restructuring that preceded this success. Moving data scientists into product teams reflects a growing understanding that AI capabilities must be tightly integrated with user needs, not siloed in engineering departments. This structural change likely contributed to the tool’s user-centric design.
The emphasis on prompt engineering and fine-tuning is also noteworthy. Many companies rush to deploy generic AI solutions, but Depop invested in customization to match its platform’s unique culture and language. This attention to detail likely explains why users are adopting the feature rather than rejecting it as impersonal or off-brand.
Looking forward, Depop’s roadmap—using AI for photo enhancement and fit prediction—suggests a comprehensive strategy to address multiple friction points in online resale, potentially setting new standards for the industry.
Why This Matters
Depop’s AI implementation represents a significant case study in how artificial intelligence can reduce friction in e-commerce and promote sustainable consumption. By automating the tedious aspects of listing creation, the company is making secondhand fashion more accessible to time-constrained consumers, directly supporting the circular economy movement.
This development is particularly noteworthy because it demonstrates practical AI application in the resale market, a growing sector valued at billions of dollars. The success of AI-powered features in increasing listing volume and quality could inspire similar implementations across other marketplace platforms, from eBay to Facebook Marketplace.
The strategic decision to move Depop’s data science team to the product side of the business highlights an important organizational trend: integrating AI capabilities directly into product development rather than treating them as separate technical functions. This approach enables faster iteration and more user-centric AI features.
For the broader AI industry, Depop’s focus on prompt engineering and fine-tuning to match platform-specific language and culture demonstrates the importance of customization in generative AI applications. Generic AI solutions often fall short; success requires tailoring technology to specific user communities and contexts.
Recommended Reading
For those interested in learning more about artificial intelligence, machine learning, and effective AI communication, here are some excellent resources: