Overview

GrowVista Bank is an EU-based retail and commercial bank. The bank ran marketing campaigns on a monthly basis, which passed offers to all of their customers from a broad set of offers designed manually. Because the bank’s customer accounts, card spending, and CRM history were distributed across multiple different systems, most of the promotional campaigns the bank ran were becoming irrelevant and resulted in a lower conversion rate. To address this, Bacancy built a personalization platform on Databricks for cross selling in banking products that combined these records into a single customer profile and selected the most relevant product for each customer in real time across the app, online banking, and the contact center. As a result, customers now receive offers matched to their own accounts, spending, and recent activity, which improved the bank’s cross-sell conversion.

Technologies Used

Salesforce
Databricks
Python
Fivetran
Apache Kafka
Power BI

Project Highlights

checkmark

Cross-selling personalization deployed across the mobile app, online banking, and contact-center channels.

checkmark

Core banking, card, digital, and CRM data unified in Databricks and governed through Unity Catalog.

checkmark

Real-time offer recommendations built on Databricks, with AI models retrained through MLflow.

checkmark

Consent, eligibility, and fairness checks applied to every recommendation, matched to GDPR & DORA rules.

The Challenges

1

Customer information was scattered across separate systems, like core banking, cards, digital banking, and the CRM. No single system showed a customer's full relationship with the bank, so every offer was based on incomplete data.

2

The marketing campaigns were designed manually; they couldn’t be personalized, and the same set of offers went to every customer each month, regardless of what they already held or needed. This resulted in low cross-sell conversion.

3

The bank's core systems were built on COBOL and didn't connect easily with modern APIs. Connecting a new AI recommendation engine to them took custom middleware, which added significant cost and time to the build.

4

The legacy systems couldn't keep up with changing data-privacy rules like GDPR and lacked modern encryption. Centralizing customer data for personalization made it harder to stay compliant and left that data more exposed to security threats.

Solutions by Bacancy

1

Bacancy’s data engineers built pipelines that pulled data from the core banking system, cards, digital banking, and the CRM into Databricks and combined it into a single customer view. Now the bank could see each customer’s balances, card spending, app activity, and past service in one place, and base every offer on their full history instead of a partial view.

2

Our Databricks developers replaced the manual monthly campaigns with machine-learning models that predict the next most relevant product for each customer. These models learn from the products customers already hold, how they spend, and their recent activity, so each person receives an offer suited to their own needs instead of the same generic promotion.

3

We used change-data-capture to move the bank’s data from its COBOL-based legacy systems into Databricks. The recommendation engine then ran on the data inside Databricks, so it never connected to the legacy systems directly, which helped avoid the need for expensive middleware and shortened the delivery time.

4

Bacancy’s AI consulting team built the access and security model around the personal and financial data in Databricks. Using Unity Catalog, encryption, data masking, and role-based access were implemented to control what each user could see, and personal data was classified to keep the platform compliant with GDPR.

Core Features

checkmark

A single customer-360 view shared across every model and channel.

checkmark

Real-time offer triggering from transaction and life-event signals.

checkmark

Built-in experimentation (A/B and holdout) to prove incremental revenue.

checkmark

Real-time offer-decision API serving the app, online banking, and contact-center agents.

No. of Resource

07

No. of Resource

Time Frame

Sept 2025–Jun 2026

Time Frame

Project Snapshot

GrowVista Bank Cross Selling Personalization Dashboard

Outcomes

3.2x higher cross-sell conversion than the bank's previous monthly campaigns.

28% increase in revenue per active customer, from more relevant and better-timed offers.

Less than 200ms taken to choose and deliver an offer, down from a monthly batch cycle.

A drop of 34% in opt-out rate, as customers stopped receiving irrelevant, repeated promotions.

Four disconnected systems consolidated into a single customer view, retiring the old data silos.

The platform passed fair-lending and model-risk review and went live with 0 compliance issues.

Technical Stack

Source Systems Salesforce
Batch ingestion (CDC) Fivetran
Streaming ingestion Apache Kafka
Lakehouse storage Delta Lake on Databricks
Transformation & pipelines Delta Live Tables (Lakeflow)
Feature management Databricks Feature Store
Data Modeling XGBoost
MLOps MLflow
Real-time serving Databricks Model Serving
Evaluation & monitoring Mosaic AILakehouse Monitoring
Data Governance and Security Unity Catalog
Experimentation A/B testing & holdout framework on Databricks
BI & reporting Power BI
Orchestration Databricks Workflows
CI/CD GitGitHub Actions

Experience With Bacancy

Optimize Your E-Commerce Workflow for Maximum Efficiency

iconWatch the video

Fintech Platform Scalability & Growth Optimization

iconWatch the video

Legacy Code Upgrades for Modern User Experience and Enhanced Data Analytics

iconWatch the video

Empowering Retail Marketing With Seamless User Experiences

iconWatch the video

Migrating To Laravel From Core PHP With Twilio Integration

iconWatch the video

Upgrading PHP Laravel version while scaling the Admin Dashboard

iconWatch the video

Optimizing CRM Performance With Data Management and Salesforce Integration

iconWatch the video

AI Solution Engine for Improved Performance and Error Handling

iconWatch the video

Enhancing Managed IT Services With Streamline Automation Process

iconWatch the video

SEO and Bulk Data Upload Solutions For An E-Commerce Healthcare

iconWatch the video

Elevate Your E-Commerce with Scalable UI and Enhanced Calculators

iconWatch the video

Simplifying Annuity Rate Management for Complex Calculations

iconWatch the video

Addressing Null Safety, Code Structure Issues, and Manual Deployment Issues

iconWatch the video

Optimize Task Management With Advanced Tracking

iconWatch the video

Revolutionize Applicant Tracking: Advanced Image Optimization Techniques

iconWatch the video

Cost-Effective Media Management Solution and Resolving Data Upload Issues

iconWatch the video

Enhance Your Management Service with User-Centric Innovations

iconWatch the video

2500+ Projects Experienced Innovation with Bacancy!

Get access to an experienced team of developers and engineers from Bacancy, handpicked to ace your goals. Kickstart within 48 hours, no-risk trial.

Book a 30 min call

14+

Years of Business Experience

1458+

Happy Customers

12+

Countries with Happy Customers

1050+

Agile Enabled Employees

How Can We Help?