marketing_ops · workflow
Zalando AdTech Lab uses recurrent neural networks to predict consumer order propensity from interaction histories
Zalando's ad-tech team needed to predict consumer order behavior from sequential interaction histories, but traditional machine learning required labor-intensive manual feature engineering and produced models that were difficult to interpret for monitoring and collaborator sign-off.
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 · Consumer event history ingested
Customer histories — sequences of typed events with timestamps — are captured as input for prediction.
Tools used
recurrent neural networksLSTMs
Outcome
RNNs achieved comparable or better predictive performance to the handcrafted-feature baseline while eliminating manual feature engineering, and latent-state visualizations provided interpretable insight into consumer behavior.
Results
Volumeabout the same or better performance
Grounding & classification
Source type: technical build writeup
16 fields verified against source quotes.
predictive analyticsmetric backednamed customersource backedtools describedworkflow describedecommerceaccuracy improvementemployee productivitytechnical build writeupecommerce opsmarketing ops