How PalTech Enabled Agentic Customer Engagement for a Global Fashion Subscription Leader

Sep 29, 2026

35%
Increase in Repeat Purchases
More relevant engagement driven by affinity and behavioral intelligence.
25%
Reduction in Customer Churn
Earlier identification and intervention around changing customer risk.
1:1 + Cohort-Aware
Customer Decisioning
Individualized treatment for high-value customers with dynamic cohort intelligence at scale.

PROJECT SUMMARY

A global fashion subscription leader had significant customer signals across CRM, digital, email, social, and partner channels, but fragmented identities and campaign-centric processes limited its ability to act on changing customer behavior. 

PalTech built an agentic customer intelligence platform on Google Cloud that unified customer context, generated predictive signals, reasoned across individual and similar-customer behavior, and determined the next-best engagement through existing campaign systems. 

The platform shifted customer engagement from static segmentation and predefined campaigns toward continuously adaptive customer decisioning.

CHALLENGES

  • Fragmented customer identities and duplicate profiles across multiple engagement channels.  
  • Predictive signals such as churn and affinity were difficult to translate into timely customer actions.  
  • Static segmentation could not keep pace with rapidly changing customer behavior and intent.  
  • High-value customers required individualized treatment while broader populations needed scalable, cohort-aware personalization.  
  • Campaign teams remained heavily involved in interpreting insights and determining each intervention.  

SOLUTION

  • Established Golden Customer Profiles on BigQuery, creating persistent customer context across transactions, behavior, interactions, and campaign history.  
  • Applied Vertex AI to identify changing churn risk, affinity, propensity, engagement, and upsell signals.  
  • Introduced Signal and Decision Agents to determine when intervention was warranted and reason across both individual and similar-customer context.  
  • Used an Engagement Agent to determine the appropriate action, offer, channel, timing, and journey before activating existing CRM and campaign platforms.  
  • Fed customer responses back into the intelligence layer, enabling an Observe → Understand → Decide → Engage → Learn operating model. 

Looking to Make Your Enterprise Data AI-Ready?

The next generation of personalization is not simply about predicting customer behavior. It is about determining how the enterprise should respond. 

PalTech helps enterprises build the data, AI, and agentic foundations required to turn continuously changing customer signals into intelligent action. 

www.pal.tech | info@pal.tech 

Let’s get in touch!