Problem Statement
A leading financial institution has partnered with Abubakar Asif to implement an AI-driven, multi-channel support solution designed to deliver secure, tiered customer service across its website, mobile app, and messaging platforms. The solution must provide a seamless, personalized experience that integrates effectively with existing systems, while ensuring secure identity verification for accessing sensitive account information and performing account-related actions.
The institution categorizes its clients into three tiers:
- VIP Customers: This top tier receives the highest support level, benefiting from an advanced AI agent with unlimited interactions and priority access to skilled live agents through skill-based routing.
- Premium Customers: Premium clients receive enhanced AI assistance and a limited number of live agent interactions, balancing robust support with efficient resource allocation.
- Standard Customers: Standard clients access essential AI services only, with a required 12-hour gap between interactions to prioritize resources for higher-tier clients.
- Unverified Users: Basic AI assistance provides general information on products and services. Interested unverified users are routed to a follow-up queue, enabling efficient lead capture and allocation.
The Sunshine CRM solution should efficiently manage these service tiers, optimize resource allocation, and ensure secure, AI-enabled customer interactions that meet diverse client expectations.
Solution Requirements
- Implement AI-Powered, Multi-Channel Support System
- Develop Tier-Based Service Levels
- Integrate Robust Identity Verification Mechanism
- Optimize Resource Allocation through Skill-Based Routing
- Lead Management for Unverified Users
- Seamless Integration with Salesforce
Goal:
To create a secure, scalable, and efficient support solution that optimizes customer experience and resource use across all client tiers.
Implementation Roadmap
- Planning and Requirements Gathering
- Developing AI Actions and Agentforce Service Agent Configuration
- Setting Up Omni-Channel for Skill-Based Routing
- Creating Omni-Flow for Agent and Live Agent Routing
- Setting Up Messaging Channels
- Customer Verification Setup
- Updating Omni-Flow
- Creating Embedded Service Deployment and Setting Pre-Chat
- Testing & Deployment
Solution Summary: Implementing AI-Enabled Multi-Tier Support for a Financial Institution
To address the financial institution’s need for a secure and tier-based AI-powered customer support solution, I implemented a comprehensive strategy involving meticulous planning, development, and deployment. Here’s how I achieved this:
1. Planning and Requirements Gathering
I collaborated closely with the bank’s stakeholders to understand their specific needs for each customer tier (VIP, Premium, Standard, and Unverified Users). This included identifying actions each tier could perform and defining the service levels required for both AI and live agent interactions.
2. Developing AI Actions and Configuring AgentForce Service Agents
Using the requirements gathered, I designed custom AI actions for all tiers. These actions were integrated into AgentForce service agents and organized under relevant topics to ensure tailored assistance for each tier.
3. Setting Up Omni-Channel for Skill-Based Routing
For live human agent support, I configured Salesforce Omni-Channel with skill-based routing. This ensured that when a customer requested to connect with a live agent, the system routed them to the most qualified human representative based on their needs.
4. Creating Omni-Channel Flows for AI and Live Agent Routing
I developed an Omni-Channel flow to handle routing for both AI service agents and live agents. Using a pre-chat form, I identified customer tiers and directed them to the appropriate AI or human agent based on tier-specific parameters.
5. Setting Up Messaging Channels
I established messaging channels for web, mobile apps, Facebook, and WhatsApp. Each channel was assigned a corresponding Omni-Channel flow to manage interactions seamlessly across platforms.
6. Customer Verification and Tier Selection
For customer verification, I utilized custom parameters in the messaging channel. These parameters were mapped within the embedded service deployment to identify users. If no user record was found, the system classified the individual as a new customer and routed them accordingly.
7. Updating Omni-Channel Flow with Custom Parameters
I enhanced the Omni-Channel flow by integrating the custom parameters from the messaging channels. These parameters were used to determine customer tiers dynamically and ensure accurate routing to AI agents or live agents.
8. Creating Embedded Service Deployment and Pre-Chat Setup
I created embedded service deployments and designed pre-chat forms. These forms captured critical customer information and mapped it to the custom parameters configured in the messaging channels.
9. Testing and Deployment
I conducted rigorous testing across all channels and tiers to validate functionality, ensuring secure and efficient service delivery. After confirming the solution met the bank’s requirements, I deployed it into production.
Key Outcomes:
- Personalized Tier-Based Support: Ensured each customer received service aligned with their tier, improving satisfaction and operational efficiency.
- Seamless Multi-Channel Integration: Provided consistent support across web, mobile apps, and messaging platforms like Facebook and WhatsApp.
- Secure and Scalable Solution: Incorporated robust customer verification and dynamic routing to manage resources efficiently and securely.
- Optimized Human-Agent Support: Enabled skill-based routing to connect customers with the most suitable agents, ensuring faster issue resolution.
This comprehensive solution exemplifies a robust, scalable approach to delivering AI-enabled, multi-tiered customer support in the financial sector.
