ecommerce_ops · workflow
DoorDash builds DashCLIP: a multimodal embedding framework for CPG ad ranking and retrieval
DoorDash's ads quality team relied on categorical and numerical features that failed to capture the semantic richness of product catalogs or a deeper understanding of user interests, limiting performance on ranking and retrieval tasks.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · Product catalog data ingestion
Roughly 400,000 products including titles, images, and aisle categories were curated for continual pre-training and evaluation.
Tools used
DashCLIPBLIP-14MGPTLLMsCLIPBLIPFLAVA
Outcome
DashCLIP outperformed all baseline models by significant gains on retrieval tasks, improved ranking metrics, increased engagement rates and revenue for sponsored product ads, and was deployed to serve 100% of traffic.
What failed first
Off-the-shelf models lacked e-commerce domain specificity and frequently failed on short but specific queries; historical engagement data introduced position and selection bias into training labels.
Results
Volumeincreased engagement rates for most of the top queries and categories
Cost replaceddriving more revenue for sponsored products ads
Grounding & classification
Source type: technical build writeup
31 fields verified against source quotes.
computer visionpersonalizationrecommendation systemproduct catalogbuilder submittedfailure mode describedhuman review describedmetric backednamed customerproduction runtime claimedsource backedtools describedworkflow describedecommerceaccuracy improvementrevenue increasethroughput increasetechnical build writeupecommerce opsmarketing ops