Viston helped us deploy AI-powered personalization, dynamic pricing, and automated content generation to increase conversions during peak sales while maintaining predictable infrastructure costs.

Business Challenge

Conversion rates were stagnant at 1.6%, Average Order Value (AOV) remained at ₹1,980, and Customer Acquisition Cost (CAC) continued to rise due to high mobile cart abandonment.

Our Approach and Solution

Viston developed a Recommendations API supporting both session-based recommendations and user cold-start scenarios. A dynamic pricing engine was implemented with margin floor guardrails to protect profitability, while AI-powered product content and SEO descriptions were generated using brand-safe prompts.

AI-Powered Recommendations

Built an intelligent recommendation engine to deliver personalized product suggestions for both new and returning customers.

Dynamic Pricing

Implemented AI-driven pricing optimization based on pricing elasticity while maintaining predefined profit margin thresholds.

AI Content Generation

Automated product descriptions and SEO content creation using Large Language Models (LLMs) with built-in brand safety and toxicity checks.

AI Customer Support

Deployed a multilingual AI chat assistant to handle routine return requests, sizing inquiries, and customer support while escalating complex cases to human agents.

AI Applications and Benefits Delivered

Personalized Recommendations

Sequence models and embedding-based recommendation engines to improve customer engagement and product discovery.

Dynamic Pricing Intelligence

AI-powered pricing elasticity estimation for optimized pricing decisions.

AI Content Automation

LLM-based content generation with toxicity detection and brand compliance validation.

Conversational AI Support

AI-powered customer support assistant for faster response times and improved shopping experiences.

Business Benefits

• Higher Average Order Value (AOV)
• Improved conversion rates
• Reduced Customer Acquisition Cost (CAC)
• Faster product content operations

Cost of Implementation

USD 40,000, including Vertex AI usage, feature store implementation, and six months of observability support.

Time to Implement

10 weeks from solution design to full production rollout.

Tools and Technologies Used

• Google Cloud Vertex AI
• BigQuery
• Feature Store
• Recommendations AI
• Cloud Run
• Redis
• LangChain
• Cloud Armor
• Firebase A/B Testing
• Datadog

Quantitative Outcomes

• 18% increase in Average Order Value (AOV) to ₹2,336
• 42% increase in conversion rate to 2.3%
• 21% reduction in Customer Acquisition Cost (CAC)
• 5× increase in content production throughput with editorial approval in under 24 hours
• 24% increase in revenue per session

Key Performance Indicators (KPIs) Tracked

• Recommendation Click-Through Rate (CTR)
• Revenue per session
• NDCG@10
• Cart abandonment rate
• Customer Acquisition Cost (CAC)
• Customer Satisfaction Score (CSAT)
• LLM toxicity rate

Pre- and Post-Implementation Metrics

• Conversion Rate: 1.6% → 2.3%
• Cart Abandonment Rate: 71% → 58%
• Product Page Publishing Time: 3.5 days → 16 hours
• Revenue per Session (A/B Test): +24% uplift (p<0.05)

Stakeholder Quotes

Chief Growth Officer

“The personalization engine felt ‘native’ to our brand. Editorial control never took a back seat.”

Regulatory and Compliance Considerations

• India DPDP Act 2023 compliance
• PCI DSS scope isolation
• Content safety filters
• Opt-out mechanisms
• Data minimization controls