Skip to content

Latest commit

 

History

History
181 lines (140 loc) · 8.21 KB

customization.md

File metadata and controls

181 lines (140 loc) · 8.21 KB

Customization

Include Your Business Data

In the /data folder, you'll find the following files:

  • agency-description.md: Contains information about your agency or business, including name, description, and contact details. Replace this with your own agency or business details.
  • case_studies: Includes past case studies that will be referenced in reports, emails, and interviews. Update this file with your own case studies to personalize the generated content.

Ensure that these files are updated to reflect your own company's information and the specific details you want to highlight in your outreach and reports.

Customizing CRM Integrations

Overview

To integrate your custom CRM with the LeadLoaderBase class, you need to inherit from LeadLoaderBase and implement two methods:

  • fetch_records: Fetches the leads matching the given status.
  • update_record: Updates the lead record with new fields.

By default, the available statuses used in the LeadLoaderBase class are:

available_statuses = ["NEW","CONTACTED"]

However, these statuses can be customized to match the specific lead status values in your CRM.

Steps to Add a Custom CRM Integration

  1. Create a New Class for Your CRM Inherit from LeadLoaderBase and implement the two abstract methods: fetch_records and update_record.

    class CustomCRMLeadLoader(LeadLoaderBase):
    def __init__(self, api_key, custom_parameter):
        # Initialize any API clients or configurations here
        self.api_key = api_key
        self.custom_parameter = custom_parameter
 
 def fetch_records(self, lead_ids=None, status="NEW"):
     """
     Fetch leads from your CRM.
     - If `lead_ids` is provided, fetch those specific leads.
     - Otherwise, fetch leads matching the given status.
     """
     if lead_ids:
         leads = []
         for lead_id in lead_ids:
             lead = crm_api.get_lead(lead_id)  # Replace with your CRM API call
             if lead:
                 # Extract all fields provided by database
                 lead = {"id": record["id"], **record.get("fields", {})}
         return leads
     else:
         # Fetch leads matching the status
         records = crm_api.get_leads_by_status(status)  # Replace with your CRM API call
         return [
             {"id": record["id"], **record.get("fields", {})}  # Extract all fields provided by database dynamically
             for record in records
         ]

 def update_record(self, lead_id, updates: dict):
     """
     Update a single lead record in your CRM.

     Args:
         lead_id (str): ID of the lead to update.
         updates (dict): Dictionary of fields to update.

     Returns:
         dict: Updated lead record.
     """
     lead = crm_api.get_lead(lead_id)  # Replace with your CRM API call
     if not lead:
         raise ValueError(f"Lead with ID {lead_id} not found.")
     updated_lead = crm_api.update_lead(lead_id, updates)  # Replace with your CRM API call
     return updated_lead

  2. Customizing Lead Statuses By default, the class uses the following statuses:

    available_statuses = [
    "NEW",
    "UNQUALIFIED",
    "ATTEMPTED_TO_CONTACT"
]

    If your CRM uses different statuses or additional ones, you can customize this list to match your system. For example, if your CRM uses statuses such as "IN_PROGRESS" or "CONTACTED", you can update the available_statuses like this:

    available_statuses = [
    "NEW",
    "IN_PROGRESS",
    "CONTACTED"
]

    Make sure to update the fetch_records method to reflect these changes. For example, if you want to fetch leads with a status of "IN_PROGRESS", you can call fetch_records(status_filter="IN_PROGRESS").

Customizing CRM Field Names

Overview

By default, in the nodes function uses the fields names used in my own database. However, different CRMs or database schemas may use different field names or additional fields that need to be handled.

Steps to Customize

  1. Identify the Fields in Your CRM Each CRM or database may have different field names or additional fields. Start by identifying which fields you need to map from your system to the CRM. This may include:

    • Custom field names (e.g., lead_score instead of Score)
    • Additional fields (e.g., custom_field_1, custom_field_2)

  2. Modify the nodes::get_new_leads Function The get_new_leads function fetches lead data from a database and structures it using specific field names (e.g., "First Name", "Last Name", "Email", etc.). If your database uses different field names, you will need to modify this function accordingly.

    For example, if your database uses first_name instead of First Name, update the code like this:

     def get_new_leads(self, lead_ids):
     ...

     # Fetch new leads using the provided loader
     raw_leads = self.lead_loader.fetch_records()
     
     # Structure the leads
     leads = [
         LeadData(
             id=lead["id"],
             name=f'{lead.get("first_name", "")} {lead.get("Last Name", "")}',
             email=lead.get("Email", ""),
             phone=lead.get("Phone", ""),
             address=lead.get("Address", ""),
             profile="" # will be constructed
         )
         for lead in raw_leads
     ]

     ...

    IMPORTANT: The provided field names are based on the structure I use for my own database. If your database uses different field names, just update the code to reflect that.

  3. Modify the nodes::update_CRM function The new_data dictionary in the update_CRM function is where the lead data is prepared before updating the CRM. You can modify this dictionary to include your custom fields.

    For example, if your CRM has a custom field custom_lead_score instead of Score, you can modify the dictionary like this:

    new_data = {
    "Status": "ATTEMPTED_TO_CONTACT",  # Set lead to attempted contact
    "Custom Lead Score": state["lead_score"],  # Updated field name
    "Analysis Reports": state["reports_folder_link"],
    "Outreach Report": state["custom_outreach_report_link"],
    "Last Contacted": get_current_date(),
    "Custom Field 1": state["custom_field_1"],  # Example of an additional field
}

Notes:

  • Field Names: Always check the database to ensure that the field names you're using match the database schema (names are case-sensitive).

Customizing Prompts

Overview

In the prompts.py file, you will find predefined prompts used for various tasks in your outreach automation. These prompts can be customized to better fit your specific needs, whether for qualifying leads, generating reports, personalizing outreach emails, or preparing interview questions. This guide explains how to update and modify these prompts.

1. Qualifying Leads

The prompt used for qualifying leads can be updated to reflect different criteria, depending on the business requirements. You can modify this prompt to suit your specific qualification process and the fields available in your CRM.

2. Generating Reports

The report generation prompt can be customized to match different report formats, structures, or analysis points. You can add or remove sections, change the data fields, or adjust the language to fit your needs.

3. Personalized Lead Outreach Email Template

The lead outreach email template can be modified to reflect your personalized messaging. This includes updating the language, structure, or including any unique data that you want to add to the email. You can tailor the subject lines, email body, or call-to-action based on your audience and goals.

4. Generating Interview Preparation Questions

The interview preparation script and questions can be customized by modifying the following prompts:

  • GENERATE_SPIN_QUESTIONS_PROMPT: Tailor the questions to focus on specific aspects of the lead’s needs, role, or challenges.
  • WRITE_INTERVIEW_SCRIPT_PROMPT: Adjust the script to focus on the key areas you want to assess during the interview.