How a leading e-commerce company in its market is using GenAI to improve customer experience and scale operations
- Denis Pesa
- Feb 26
- 4 min read
Personalization in customer service Automation of the recommendation process Scalability with artificial intelligence
AI generated summary:
A well-established company in the corporate gifts market, operating three online platforms, faced challenges in personalizing and scaling its service due to the complexity of the sector. To optimize the first service and recommend products more accurately, BlueMetrics developed a solution based on GenAI. The technology enriched product category data, created a structured knowledge base and implemented a contextual virtual assistant.
Overview
Our client is a well-established company in the corporate gifts market, operating three online platforms that connect suppliers and buyers. Highly competitive, this sector is characterized by high complexity in decision-making, due to the extensive variety of products and specificities of each demand.
To stand out in a constantly evolving market, the client needed a solution capable of combining personalization, agility and precision in the customer experience, in addition to scaling its operations without increasing proportional costs. Based on this context, BlueMetrics was called to propose a solution capable of bringing significant improvements to the customer journey.
“We are experts in developing real solutions for real problems,” says Gabriel Casara, CGO of BlueMetrics. “Therefore, this type of challenge is very much in line with both our working method and our solution offering.”
Problem: How to optimize initial service and product category recommendations?
This client’s operations faced significant challenges that compromised efficiency and customer experience. Service was limited to business hours, which restricted support to customers outside of business hours. In addition, there was a high reliance on individual agent knowledge, resulting in a manual process of interpreting requests and directing them, often leading to delays and errors.
Among the technical limitations, it was identified that the category data had little semantic content and a lack of systematization regarding purposes and events appropriate for each category, which made it difficult to adopt an intelligent recommendation system.
According to Diórgenes Eugênio, Head of GenAI at Bluemetrics, “The big difference in the virtual assistant project for this e-commerce site is precisely the creation of the knowledge base through the use of LLM models. This approach allowed us to deliver much more context to the assistant. At first, we had very little semantic information about the categories, and we finished the project with an automated pipeline that processes all the content coming from the e-commerce site and semantically and contextually improves the data to serve as a source of reference for the virtual assistant. This project was a pioneer within Bluemetrics in terms of the use of LLMs for data enrichment.”
From a business perspective, delays in first-time customer service negatively impacted customer satisfaction. Furthermore, the overload of staff during seasonal periods, such as Christmas and the end of the year, further aggravated the situation, leading to imprecise directions, rework and loss of business opportunities. These bottlenecks generated a series of direct impacts on the business:
Customer dissatisfaction with high response times;
Unintentional favoritism of certain suppliers;
Limitation in business growth due to the manual service model.
Faced with these challenges, the client needed a scalable, impartial solution capable of providing 24/7 service, reducing response time and democratizing access to supplier options. Therefore, BlueMetrics developed an intelligent solution to automate and optimize the initial service process.
The solution: GenAI for personalization and scalability
To address the challenges identified, the client implemented a robust solution based on Artificial Intelligence, structured around three main pillars, as we will see below:
Data enrichment
This process begins with the processing of XML data extracted from customer platforms, using Amazon Bedrock LLM models to enrich product category descriptions. In addition, relevant context about events and appropriate purposes for each category is added, resulting in a rich and highly structured knowledge base that serves as the foundation for other functionalities.
Smart knowledge base
The enriched information is converted to PDF files and stored in a vector database optimized for semantic search. This architecture ensures not only efficient search but also continuous updating of the data, maintaining the relevance and accuracy of the information over time.
Contextual virtual assistant
This assistant is designed to interact naturally with customers, understanding their context and specific needs. Using Information Retrieval (IRA) techniques, it offers relevant and unbiased recommendations, suggesting product categories accurately and appropriately for each situation.
Integrated, these components resulted in an innovative and effective solution, allowing the client to optimize initial service, reduce operational bottlenecks and provide a more personalized and satisfactory shopping experience for its customers.
How about developing a solution like this for your company?
Results:
The implementation of the virtual assistant based on Artificial Intelligence brought a series of benefits to the client, translating into concrete and immediate financial results.
Operational Benefits
24/7 service, eliminating dependence on business hours;
Reduction in initial waiting time for service;
Standardization in the recommendation process;
Unlimited simultaneous service capacity;
Reduction of manual workload for staff.
Technical benefits
Semantically enriched knowledge base;
Scalable and flexible architecture;
Ease of incorporating new LLM models;
Simplified knowledge base maintenance.
Customer Benefits
Instant responses to requests;
More precise and contextualized recommendations;
Unbiased category suggestions;
Better experience in the purchasing journey;
Greater assertiveness in choosing products.
Technologies used
The solution developed for this e-commerce client was designed using AWS technologies, including:
AWS Services
OpenSearch
Bedrock
Lambda
CloudWatch
S3
Amplify
Cognito
StepFunction
Languages, Libs and Frameworks
Python
Streamlit
Fast API
Conclusion
The implementation of the GenAI-based solution has enabled this e-commerce player to scale its operations and significantly improve customer experience, further consolidating its position in the corporate gifts market. With a robust, scalable and highly personalized system, the company is now prepared to meet growing demand, maintaining quality and assertiveness as competitive differentiators.
How about creating a case like this for your company? Let's schedule a call?
Discover some Use Cases .
About BlueMetrics
BlueMetrics was founded in 2016 and has already delivered more than 160 successful solutions in the areas of Data & Analytics, GenAI and Machine Learning for more than 70 companies in the United States, Brazil, Argentina, Colombia and Mexico. It has its own methodology and a multidisciplinary team focused on delivering solutions to real challenges in the business world.
Comentários