Fine-tuning LLM

Case Study: Finequities & Bitext Copilot – Redefining the New User Journey in Social Finance

In the dynamic world of social investments, where user-platform interaction plays a crucial role in the success of customer acquisition and retention, Finequities, a leading social investment app, faced the challenge of optimizing its onboarding process to enhance user experience. The solution came through an innovative collaboration with Bitext, a pioneer company in dialogue technologies and natural language processing, which introduced the Bitext Copilot, a proactive assistance tool designed to radically transform the way users interact with the Finequities platform.

Challenge

The traditional onboarding process of Finequities, based on static forms, presented several challenges: it was cumbersome, not intuitive, and couldn’t handle complex requests or personalize the user experience. This resulted in a significant barrier to acquiring new users and negatively affected customer satisfaction.

Solution

Bitext introduced the Copilot, a natural language interface that replaces static forms with a conversational, proactive, and highly personalized user experience. This change not only simplified the onboarding process but also made it more interactive and capable of resolving queries in real time, offering significant advantages over traditional methods:
  • Proactive Assistance and Intuitive Interaction: The Copilot anticipates and responds to user questions naturally, guiding them through the process with ease.
  • Advanced Personalization: Tailors its responses and suggestions to the specific preferences and needs of each user.
  • Handling of Complex Requests: Understands and manages detailed requests efficiently without the need for detailed manual interaction.
  • Flexibility in Changes and Requests: Allows users to make modifications easily, offering alternatives and proactively managing changes.
  • Improved Accessibility: Its ability to assist with voice commands makes it easier for people with visual impairments or other limitations.
  • Cognitive and Adaptive Use: Learns from each interaction to improve future responses and suggestions.
  • Dynamic Response to Queries: Adjusts in real time to user questions, ensuring relevance and accuracy.

Implementation and Results

The integration of Bitext’s API enabled Finequities’ development team to fully customize the onboarding-specific Copilot in record time: just two weeks. This was made possible by direct access through a single API Key to GPT LLM, allowing for easy implementation of a frontend with dialogue management and access to onboarding-specific pre-trained models, an exclusive advantage of Bitext not available in APIs from competitors like OpenAI or Microsoft Azure Open AI.

In the following video, you can see Finequities Copilot in action during the onboarding process of a new user:

Key Benefits

  • Rapid and Efficient Integration: With Bitext’s API, Finequities achieved full integration in just two weeks.
  • Reduced Cost and Time: Access to pre-trained models allowed Finequities to save on costs and time, eliminating the need for custom training data preparation.
  • Transformed User Experience: The onboarding process became more interactive, intuitive, and personalized, significantly improving customer satisfaction.
OpenAI Chat API OpenAI Assistant API Bitext Copilot API
Fine-tuning X X
Pre-train Fine-tuning models X X
Fast X X
Structured Data X X X
Conversation Handling X X
RAG X X
API Integration Complexity Medium Hard Easy

Conclusion

The collaboration between Finequities and Bitext has set a new standard in onboarding experience in the social investment industry. By replacing static forms with a proactive conversational interface, Finequities has not only improved the efficiency of the onboarding process but has also enriched the user experience, facilitating a clearer path to customer engagement and retention. This case study highlights the power of technological innovation applied to improving user interactions, marking a milestone in the evolution of platforms.
Pedro Hernandez

Recent Posts

Integrating Bitext NAMER with LLMs

A robust discussion persists within the technical and academic communities about the suitability of LLMs…

2 days ago

Bitext NAMER Cracks Named Entity Recognition

Chinese, Southeast Asian, and Arabic names require transliteration, often resulting in inconsistent spellings in Roman…

2 weeks ago

Deploying Successful GenAI-based Chatbots with less Data and more Peace of Mind.

Customizing Large Language Models in 2 steps via fine-tuning is a very efficient way to…

6 months ago

Any Solutions to the Endless Data Needs of GenAI?

Discover the advantages of using symbolic approaches over traditional data generation techniques in GenAI. Learn…

7 months ago

From General-Purpose LLMs to Verticalized Enterprise Models

In the blog "General Purpose Models vs. Verticalized Enterprise GenAI," the focus is on the…

8 months ago

Automating Online Sales with Proactive Copilots

Automating Online Sales with a New Breed of Copilots. The next generation of GenAI Copilots…

11 months ago