Keyword limits and manual workflows bottlenecked sales. Semantic search raised purchase amounts 14.2% and halved listing creation time.
The digital strategy arm of a major Japanese apparel group operates an e-commerce platform handling over 100,000 monthly product searches.
Conventional keyword-based search failed to handle spelling variations or understand user intent, preventing customers from finding relevant items....
“従来の検索エンジンでは、商品名や商品説明に記載されたキーワードのみに依存していたため、表記のゆらぎなどがあった際に、顧客が求める商品にたどり着けないという課題がありました。Vertex AI Search for commerce は、Google 検索のように意味を理解したセマンティックな検索が導入されているので、キーワードでマッチしなかったとしても、意図どおりの商品にたどり着けることが増えました。”
Onward Digital Lab's Product search and content is part of this use case:
Digital solutions and e-commerce for the fashion and apparel industry.
Cloud computing services, AI infrastructure, and data analytics platforms for enterprises.
Implemented Vertex AI Search for commerce, Gemini 2.5 Flash Image, and Veo for Onward.
Related implementations across industries and use cases
Exact-match search missed synonyms. AI now maps intent—linking "cooling goods" to inventory—turning vague queries into sales.
Keyword search missed nuance. AI now decodes intent—like "wet" meaning waterproof—to curate items like an in-store expert.
Building 1,500 seasonal product pages relied on manual image matching. Now, AI pairs assets to assemble layouts and generate social ads.
Messy ingredient data broke text search, capping matches at 20%. AI semantic search now maps free-form inputs to unlock recipe imports.
Merchandisers manually updated rules weekly to decode trade slang. Now, AI continuously trains on live data to process complex queries.
Building an air conditioner every seven seconds outpaced manual assembly. Now, 2,000 AI agents run three automated factories 24/7.
Manual workflows delayed global campaigns for weeks. Now, local teams use GenAI to instantly draft localized, regulatory-compliant copy.
Shoppers built lists item-by-item. Now, they build full baskets in one prompt, while staff automate internal workflows.
Data scattered across systems forced generic content. Now, AI agents surface personalized stats and videos via chat during live matches.
Keyword limits and manual workflows bottlenecked sales. Semantic search raised purchase amounts 14.2% and halved listing creation time.
The digital strategy arm of a major Japanese apparel group operates an e-commerce platform handling over 100,000 monthly product searches.
Conventional keyword-based search failed to handle spelling variations or understand user intent, preventing customers from finding relevant items....
“従来の検索エンジンでは、商品名や商品説明に記載されたキーワードのみに依存していたため、表記のゆらぎなどがあった際に、顧客が求める商品にたどり着けないという課題がありました。Vertex AI Search for commerce は、Google 検索のように意味を理解したセマンティックな検索が導入されているので、キーワードでマッチしなかったとしても、意図どおりの商品にたどり着けることが増えました。”
Onward Digital Lab's Product search and content is part of this use case:
Digital solutions and e-commerce for the fashion and apparel industry.
Cloud computing services, AI infrastructure, and data analytics platforms for enterprises.
Implemented Vertex AI Search for commerce, Gemini 2.5 Flash Image, and Veo for Onward.
Related implementations across industries and use cases
Exact-match search missed synonyms. AI now maps intent—linking "cooling goods" to inventory—turning vague queries into sales.
Keyword search missed nuance. AI now decodes intent—like "wet" meaning waterproof—to curate items like an in-store expert.
Building 1,500 seasonal product pages relied on manual image matching. Now, AI pairs assets to assemble layouts and generate social ads.
Messy ingredient data broke text search, capping matches at 20%. AI semantic search now maps free-form inputs to unlock recipe imports.
Merchandisers manually updated rules weekly to decode trade slang. Now, AI continuously trains on live data to process complex queries.
Building an air conditioner every seven seconds outpaced manual assembly. Now, 2,000 AI agents run three automated factories 24/7.
Manual workflows delayed global campaigns for weeks. Now, local teams use GenAI to instantly draft localized, regulatory-compliant copy.
Shoppers built lists item-by-item. Now, they build full baskets in one prompt, while staff automate internal workflows.
Data scattered across systems forced generic content. Now, AI agents surface personalized stats and videos via chat during live matches.