AI Broking Assistant: Project Initiation Report

This section outlines the client's overarching vision for the AI Broking Assistant project. It covers the primary objectives, the desired impact on their operations and the broader industry, and the long-term aspirations for the product. Understanding this vision is key to aligning development efforts with strategic goals.

The primary goal is to develop an AI-powered broking assistant to enhance efficiency within an insurance brokerage. This assistant would automate repetitive tasks, freeing up human brokers to focus on more complex client interactions and advisory roles.

The client expressed excitement about the "endless capabilities" and potential for such technology to be a "significant disruption."

This section details the core problems the client aims to solve and the specific functionalities discussed for the AI assistant. It includes information the AI would need to access and actions it would need to perform, primarily focusing on interactions with the CRM and email systems. The chart below visualizes the mention frequency or implied importance of these capabilities.

Problem Statement:

Brokers are burdened with numerous repetitive tasks, notably fetching documents and handling client communications. The AI assistant is envisioned to alleviate this by automating these processes.

Key AI Capabilities Discussed:

• Read from CRM: Access customer profiles, support tickets/cases, communication logs, knowledge base articles, internal documents, and product information to understand context and fulfill requests.
• Write/Update CRM: Log actions taken by the AI, update ticket statuses, and add notes to client records.
• Fetch Documents: Retrieve specific documents from the CRM based on broker or automated triggers.
• Email Automation: Send documents and other communications to clients on behalf of brokers.

Capability Focus (Mention Frequency):

This chart illustrates the relative emphasis placed on different AI capabilities during the initial discussion.

To provide a concrete example of automation, the client shared a specific task: generating a "Debtors Ageing Schedule Report." This section breaks down this task and its proposed extension, illustrating a practical application of the AI assistant. This use case highlights how the AI can handle multi-step processes and follow-up actions.

A key example task involves generating a report of clients who owe money to the company. The client provided a ScribeHow link detailing 7 steps for this process.

Task: Generate Debtors Ageing Schedule Report

Show Detailed Steps (7)

1. Log in to the custom CRM system.
2. Navigate to the 'Reports' or a similar section.
3. Select 'Financial Reports' or a specific category for debtor reports.
4. Choose the 'Debtors Ageing Schedule' report type.
5. Configure report parameters (e.g., date range, aging buckets like 30, 60, 90+ days).
6. Initiate report generation by clicking a 'Generate' or 'Run Report' button.
7. Review the generated report and optionally export it (e.g., as PDF or CSV).

Automated Follow-up Actions:

Beyond generating the report, the AI assistant would then be prompted to:

• Download a statement of account for each overdue client identified in the report.
• Compose and send personalized email reminders to these clients, attaching their statements, to remind them that payment is overdue.

This entire workflow would be driven by each broker, allowing them to initiate and oversee these automated processes.

This section provides background on the client's existing technical infrastructure and operational environment. Understanding these aspects, such as the nature of their CRM and the client's technical proficiency, is crucial for designing a compatible and user-friendly AI solution.

Existing Systems:

• Custom CRM: The brokerage uses a custom-built CRM. This system is central to their operations.
• Insurer Partner Connectivity: The CRM connects to "a few" insurer partners, suggesting API integrations or data exchange mechanisms might exist.
• Email Integration: A requirement for the AI to connect to their email system to send communications on brokers' behalf.

Client & Team Profile:

• Client Technical Background: The client (Daniel) stated, "I am certainly not a developer... I don’t quite understand the tech side of things so you will have to guide me through that process." This highlights the need for clear communication and potentially a user-friendly interface for the AI's controls.
• Client Role: Manages multiple businesses and will provide ideas for additional tech needs.
• Development Team (Renesis Tech): Identified as "a team with devs specializing in specific roles."

This section summarizes the logistical aspects of the collaboration discussed, including meeting arrangements and information exchange. It outlines the immediate next steps proposed to move the project forward, such as a more detailed requirements gathering call.

Meeting Logistics:

• Timezones: Client is GMT+10, Development Team (Renesis Tech) is GMT+5.
• Proposed Meeting: Scheduled for 3:30 PM client's time (Australian time), for 30-45 minutes.
• Communication Channels: MS Teams is the official channel for Renesis Tech. Freelancer calling feature (with screenshare) was also suggested as an option.

Information Exchange & Proof of Concept:

• Request for Documentation: The development team requested any existing documents containing requirements to prepare for the call effectively.
• Potential Proof of Concept (PoC): Reviewing documents in advance might allow the team to build a PoC to present during the call.
• Client Contribution: The client shared the ScribeHow link for the Debtor Ageing Schedule task as an initial piece of information and expressed willingness to "keep providing ideas on the additional tech needed."

The immediate next step is the scheduled meeting to discuss the project in more detail and review any provided documentation.