Etsy's AI: Algotorial Magic Unleashed

How Etsy used Google Cloud AI to personalize 90M shopping experiences and boost listings 80x

AI of the Tiger Newsletter

🐯 AI OF THE TIGER 🐯

September 6, 2025

TL;DR:Etsy used Google Cloud's AI arsenal to transform its marketplace into a personalization powerhouse—boosting listings per theme by 80x, increasing SEO visits by 5%, and delivering custom shopping experiences to all 90 million buyers. This is "algotorial curation" at its finest.

🎯 AI IN ACTION

🎯 Background & Mission

Picture this: You're running a marketplace with 130 million unique items from 5 million sellers, and you need to connect each of your 90 million shoppers with exactly what makes their heart skip a beat. That's Etsy's daily reality—and their "Keep commerce human" mission just got a serious AI upgrade.

"There's only one marketplace where you know you're buying from another real person on the other side, and that's Etsy. We're not trying to replace that human connection. We want to amplify and scale that connection with AI." - Mahir Yavuz, Senior Director of Engineering for Search, Recommendations, and Ads at Etsy

🔗 Enriching Buying and Selling with Multimodal AI

Etsy's sellers don't just upload photos—they share stories through images, videos, and text. The challenge? Making sense of this multimodal treasure trove at scale.

"Our sellers provide us images, videos, and textual information, and we want to use everything possible. The ability to go multimodal has become critical to providing great shopper experiences."

Early AI experiments in inventory management revealed the power of high-quality, structured data. This success inspired Etsy to expand AI's role, leading to further innovations in inventory management and discovery. Multimodal AI enables Etsy to analyze and understand listings more deeply, improving search relevance, recommendations, and the overall buyer journey.

💡 Understanding Shopper Intent for Instant "Click" Moments

Here's where things get interesting. Sometimes shoppers arrive at Etsy knowing exactly what they want. Other times? They're just browsing, hoping something will spark joy.

"In the case a buyer is unclear what they want, we pick the best items for you using AI."

Etsy's approach is two-pronged: Expert curators create seed collections that capture trends and styles, then AI amplifies this work exponentially.

"Our experts cannot create millions of different styles or listings, but they can create the seed dataset, which we can scale with AI."

Etsy also suggests "pathways" for buyers to explore inventory, built using Vertex AI and advanced ML to find semantically relevant objects. This quickly surfaces the best of Etsy in response to each visitor's choices. As buyers interact and search, their intent becomes clearer, allowing Etsy's AI to refine recommendations in real time.

"It doesn't matter whether the buyer is looking to find something only on Etsy. It's also important to give them the best of the information anywhere on the web."

☁️ AI & Google Cloud Solution

Etsy's AI transformation runs on Google Cloud's heavyweight lineup: Vertex AI, BigQuery, Dataflow, and Gemini models. This isn't just about throwing technology at a problem—it's about creating a foundational dataset that makes everything else possible.

"Most of Etsy's inventory management products are built on Google Cloud and use BigQuery, Dataflow, the Vertex AI Platform, and Gemini models to understand inventory and connect every buyer with the right item."

🎨 "Algotorial Curation": Human + AI at Scale

Here's Etsy's secret sauce: "algotorial curation"—a brilliant blend of human expertise and AI scalability. Human merchandising experts identify trends and create curated collections, then AI takes over to find similar items and personalize the experience for each shopper.

The result? An 80x increase in listings per theme. That's not a typo—eighty times more relevant items surfaced for each trend or style category.

🎬 See Etsy's AI Transformation in Action

Watch how Etsy keeps human connection at the heart of commerce while leveraging Google Cloud AI to create personalized experiences for millions of shoppers.

Watch on YouTube

🏗️ Deep Inventory Understanding & Dynamic Taxonomy

With 130 million items changing daily, static categorization is like trying to organize a hurricane. Etsy uses AI to dynamically understand each item's characteristics, determine optimal categorization, and assess popularity in real-time.

"I like hats. I wear hats all the time. On Etsy, we have a lot of hat shops. You can find handmade hats. Vintage hats. Custom hats."

No matter the item, Etsy's AI generates, evaluates, enhances, and regenerates data to improve accuracy and precision, surfacing the ideal item for each shopper.

AI enables this deep understanding of Etsy's inventory at immense scale.

"Some of these things were not possible before the invention of LLMs because it would cost so much money and time. Gen AI is fast, it's scalable, and it's also affordable now."

📈 Trend Detection & Cultural Relevance

Social media moves fast, and Etsy sellers move faster. When a trend emerges on TikTok or Instagram, sellers quickly create related items. Gemini models help Etsy merchandisers spot these emerging trends and classify new items before the trend even peaks.

🔍 SEO & Alt Text Optimization

Etsy used Gemini to improve alt text generation for listings, resulting in a 5% increase in SEO-driven visits and a 3% conversion lift for sellers. That's the power of AI making content more discoverable across the entire web.

📊 Quantifiable Business Results

  • 80x increase in listings per theme through algotorial curation
  • 90 million personalized shopping experiences delivered
  • 5% increase in SEO-driven visits
  • 3% conversion lift for sellers
  • 130 million items dynamically understood and categorized

⚠️ Implementation Challenges

Scaling AI across 130 million constantly changing items isn't exactly plug-and-play. The biggest hurdle? Creating that foundational dataset using Gemini models—essentially teaching AI to understand the nuanced characteristics of handmade and vintage items at massive scale.

Cost was another consideration.

"Some of these things were not possible before the invention of LLMs because it would cost so much money and time. Gen AI is fast, it's scalable, and it's also affordable now."

🎓 Lessons Learned

  • Start with the foundation: Invest in creating high-quality training data before building applications.
  • Blend human expertise with AI scale: The "algotorial" approach leverages the best of both worlds.
  • Think multimodal from day one: Images, videos, and text together tell a richer story.
  • Focus on intent, not just content: Understanding why someone is shopping matters as much as what they're looking for.

🐯 Tiger Takeaway:

Etsy cracked the code on AI-powered personalization by remembering that technology should amplify human connections, not replace them. Their "algotorial curation" approach—combining human expertise with AI scale—delivered measurable results across the board. The lesson? Don't choose between human insight and AI efficiency. Use AI to make human expertise infinitely scalable.

Sources: Google Cloud Etsy Case Study, Google Cloud AI Blog, Quotes from Mahir Yavuz

Questions or feedback? Just reply to this email—we read every message.

Want to browse past issues?Visit our website for the full newsletter archive.

Has this newsletter been forwarded to you?Click here to subscribe

AI Insights for Business Leaders

🤖

AI-Powered Newsletter

This newsletter is generated through an AI automation system featuring specialized Research, Writer, and Publisher agents. Each agent utilizes advanced tools for content discovery, analysis, and formatting. Human oversight is maintained at every step to ensure quality, accuracy, and editorial standards.