What You Will Learn

• Build a custom GPT in ChatGPT to fetch and manage emails
• Use the Actions feature and define schema for webhooks
• Create Make.com scenarios to import emails into Notion
• Set up Zapier workflows to draft/send replies and update Notion
• Customize GPT instructions and knowledge files for tone and automation

Study Guide

Introduction

Welcome to the comprehensive video course on creating an Automated GPT that sends emails directly within ChatGPT. In this course, you will learn how to build a custom GPT that can fetch real emails from your inbox, draft replies, and send them without switching between multiple applications. This automation is achieved by leveraging ChatGPT's "actions" feature in conjunction with external automation platforms like Make.com (formerly Integromat) and Zapier, as well as Notion for email storage and status tracking. This course is valuable for anyone looking to streamline their email management process, saving time and enhancing productivity.

Building a Custom GPT for Email Automation

The core concept of this course is to create a bespoke GPT within ChatGPT that extends beyond simple text generation to interact with external services for email management. This eliminates the need to switch between different applications, allowing you to manage emails directly within ChatGPT.

Example:
Imagine receiving a flood of emails daily. Instead of manually sorting, reading, and replying to each, your custom GPT fetches these emails, drafts potential replies, and lets you send them with minimal input.

Example:
Consider a scenario where you're on the go and need to manage your inbox. With this system, you can quickly process emails directly from your mobile device using ChatGPT, making it a powerful tool for busy professionals.

Utilizing the "Actions" Feature in ChatGPT

The "actions" feature in ChatGPT is pivotal for enabling the functionality of fetching and sending emails. It allows the GPT to interact with external APIs and services, which is crucial for automating the email management process.

Example:
By setting up specific actions, your GPT can trigger an API call to fetch unread emails from your inbox, process them, and present you with options for replies.

Example:
Actions can be configured to automatically update the status of an email in your Notion database once a reply is sent, ensuring seamless tracking and management.

Integration with Make.com for Email Retrieval and Notion Database Management

Make.com is used to create automations that monitor a specified email inbox, such as "all mail" or filtered for unread messages, via the Gmail module. This integration is essential for fetching emails and managing them within Notion.

Fetching Emails

Make.com scours your inbox and connects to your email collection system, pulling real emails from your inbox.

Example:
Set up a Make.com scenario that triggers every time a new email arrives, fetching details like the sender, subject, and body to be stored in Notion.

Example:
Configure filters within Make.com to only fetch emails marked as important, ensuring that your GPT focuses on high-priority communications.

Storing Emails in Notion

A Notion database acts as an "inbox" where the Make.com automation extracts key information from new emails and creates corresponding entries. The status of each email is initially set to "not reviewed".

Example:
Create a Notion database with properties such as Subject, Sender, Status, and Email Body to organize emails effectively.

Example:
Use Notion's powerful database features to filter and sort emails based on different criteria, like sender or date, for better management.

Fetching "Not Reviewed" Emails for ChatGPT

Another Make.com automation listens for a trigger via a custom webhook called by the GPT and queries the Notion database for emails with the status "not reviewed". This information is then sent back to the GPT.

Example:
Set up a Make.com scenario that triggers when your GPT requests emails, fetching only those marked as "not reviewed" to ensure you focus on pending tasks.

Example:
Incorporate a daily summary feature within ChatGPT that lists "not reviewed" emails, helping you prioritize your email management.

Integration with Zapier for Sending Email Replies and Updating Notion

Zapier is utilized to create workflows that are triggered by custom webhooks called by the GPT when the user wants to send a reply. This workflow uses the Gmail module to create a draft reply in the same thread as the original email.

Sending Draft Replies

Zapier receives reply information from ChatGPT and uses it to create a draft reply in Gmail, populated with the content generated by the GPT.

Example:
Configure a Zapier workflow that automatically sends replies drafted by your GPT, allowing you to approve or edit them before they are sent.

Example:
Set up a Zapier action that notifies you via email or a messaging app once a draft reply is ready for review, keeping you in the loop.

Updating Notion Status

The Zapier workflow also updates the status of the corresponding email in the Notion database from "not reviewed" to "replied".

Example:
Automate the update of email statuses in Notion to "replied" once a draft is sent, ensuring your database is always up-to-date.

Example:
Create a Zapier task that logs every email reply action in a separate Notion database for record-keeping and analysis.

Step-by-Step Guide to Building the System

This section provides a detailed, step-by-step process for setting up each component of the automated email system.

Creating a Notion Account and Email Inbox Database

Create a Notion account and set up an email inbox database with properties like Subject, Status, Sender Email, Sender Name, Thread ID, and Email Body.

Example:
Design a Notion template that includes all necessary properties for email management, ensuring consistency and efficiency.

Example:
Utilize Notion's sharing features to collaborate with team members on email management tasks, enhancing teamwork.

Building Automation Scenarios in Make.com

Build two automation scenarios in Make.com: one to watch for new emails and add them to Notion, and another to fetch "not reviewed" emails from Notion and send them to ChatGPT via a webhook.

Example:
Design a Make.com scenario that triggers on specific email criteria, such as emails from certain domains or with specific keywords.

Example:
Utilize Make.com's scheduling features to run automations at specific times, optimizing resource usage and performance.

Building a Zapier Workflow

Create a Zapier workflow to receive reply information from ChatGPT via a webhook, create a draft reply in Gmail, and update the email status in Notion.

Example:
Integrate Zapier with other tools like Slack or Microsoft Teams to receive notifications or alerts about email management activities.

Example:
Use Zapier's multi-step workflows to automate complex email management tasks, such as sending follow-up reminders.

Configuring the Custom GPT

Configure the custom GPT with a name, description, profile picture, conversation starters, and crucial schema for both the Make.com (fetching emails) and Zapier (sending replies) webhooks.

Example:
Custom-design your GPT's profile and conversation starters to match your brand's tone and style, enhancing user experience.

Example:
Employ schema configurations to tailor the GPT's behavior, ensuring it accurately reflects your communication preferences.

Adding Instructions to the GPT

Add instructions to the GPT to guide its behavior in fetching emails, presenting them to the user, drafting responses, and initiating the sending process via Zapier.

Example:
Incorporate specific instructions for handling different types of emails, such as customer inquiries or internal communications.

Example:
Use knowledge files to enhance the GPT's understanding of your business context, improving the relevance of its responses.

Customization and Efficiency

The system is designed to be highly customizable, allowing you to tailor the GPT's instructions and responses to your specific needs and communication style. The goal is to significantly increase the speed and efficiency of processing emails.

Example:
Customize the GPT to prioritize emails from key clients or partners, ensuring timely and relevant responses.

Example:
Leverage GPT's capabilities to automate routine email tasks, freeing up time for more strategic activities.

Community and Resources

Complete Ai Training offers a community with pre-built templates, blueprints, and step-by-step tutorials for users who want to fast-track the implementation of this and other GPT automations.

Example:
Join the AI Foundations Community to access exclusive resources and support for building and optimizing your automated email GPT.

Example:
Utilize community-shared templates to quickly deploy new automations, reducing setup time and effort.

Conclusion

By completing this course, you now have the knowledge and skills to create an Automated GPT that sends emails within ChatGPT. This powerful system allows you to efficiently manage your inbox, draft replies, and send them without leaving the ChatGPT interface. The thoughtful application of these skills can significantly enhance your productivity and streamline your email management process. Remember, the key to success lies in customizing the system to fit your unique needs and continuously optimizing it for maximum efficiency.

Podcast

Frequently Asked Questions

Introduction to the FAQ Section

Welcome to the Frequently Asked Questions (FAQ) section for the 'Video Course: Create an Automated GPT that sends Emails in ChatGPT'. This resource is designed to address common questions and concerns about automating email management using a custom GPT. Whether you're a beginner or an advanced user, this FAQ aims to provide clear and practical answers to help you navigate and implement this powerful tool effectively.

What is the core functionality of the automated GPT described in the video?

The automated GPT allows users to manage their email inbox (fetching, drafting replies, and sending) directly within the ChatGPT interface. It achieves this by connecting to external services like make.com (for accessing and processing emails) and Zapier (for sending email replies and updating a Notion database). This eliminates the need to switch between multiple applications to handle email correspondence.

How does the GPT access and interact with my email inbox?

The GPT doesn't directly access your email. Instead, it works in conjunction with make.com. A make.com automation is set up to watch your designated email inbox (e.g., all mail, unread emails). When new unread emails arrive, make.com pulls specific information (sender, subject, body, thread ID) and stores it as new entries in a Notion database. The GPT then connects to this Notion database to access the email data.

What is Notion's role in this email automation process?

Notion serves as an intermediary database where emails fetched by make.com are stored. Each email becomes a page in a designated Notion database, with properties like subject, sender, status (not reviewed/replied), and the email body as page content. This structured format in Notion makes it easier for the GPT to query, read, and process the email information compared to directly interacting with an email inbox.

How does the GPT suggest email replies, and can I customise them?

When you select an email within the GPT, it presents you with a few suggested reply options (e.g., express interest, suggest a call, decline politely). These suggestions are likely based on pre-defined instructions given to the GPT during its setup. While the video doesn't explicitly detail customisation within the GPT interface in this process, the user retains the ability to revise the suggested responses before sending them. Furthermore, the underlying instructions of the GPT could be modified to influence the style and types of suggested replies. Adding knowledge files (like FAQs or example replies) to the GPT could further enhance the customisation and relevance of the suggestions.

How are email replies actually sent using this automated system?

When you choose a reply option within the GPT, it triggers a 'post' request to a Zapier webhook. This webhook, set up in a Zapier automation, receives the necessary information (recipient, subject, body, thread ID) from the GPT. Zapier then uses this information to create a draft reply in your Gmail account, within the original email thread. The email is saved as a draft, allowing you to review and manually send it if desired, although the system can also be configured to send automatically.

How is the status of emails tracked within this system?

The system uses the 'status' property in the Notion database to track whether an email has been reviewed and replied to. Initially, when make.com pulls in a new email, its status in Notion is set to 'not reviewed'. Once a reply is drafted and sent (or marked as replied via the Zapier automation), the Zapier flow updates the status of the corresponding Notion database item to 'replied'. This ensures that the GPT primarily focuses on unread and unreplied emails.

What are the technical components required to build this automated email GPT?

Building this system requires several components: A ChatGPT Plus subscription to access the GPT builder and Actions feature. An account with make.com (or a similar integration platform) to create the automation for fetching emails and pushing them to Notion. A Notion account to create and manage the email database. A Zapier account (or another automation platform) to create the automation for sending email replies and updating the Notion database. Knowledge of how to set up webhooks in both make.com/Zapier and within the GPT's Actions configuration.

What are the potential benefits of using such an automated email GPT?

The potential benefits include: Increased Efficiency: Significantly speeds up email processing by automating tasks like fetching, drafting, and sending replies. Centralised Workflow: Allows users to manage emails without leaving the ChatGPT interface, reducing context switching. Improved Response Quality: Leveraging the language capabilities of GPT to generate well-crafted and relevant reply options. Customisation and Personalisation: The GPT can be further trained with custom instructions and knowledge to better reflect the user's communication style and handle specific email types. Organisation: The Notion database provides a structured way to track and manage email correspondence.

What is the 'actions' feature in ChatGPT, and how is it utilised in this email automation?

The 'actions' feature in ChatGPT allows the GPT to interact with external services and APIs. In this automation, it's used to connect to Make.com and Zapier, enabling the GPT to retrieve emails from Notion and send replies via Gmail. This feature is crucial for bridging the gap between the ChatGPT interface and the external platforms involved in the email management process.

Explain the role of Make.com in the process of fetching and managing emails with this GPT.

Make.com acts as the automation platform that connects the user's Gmail inbox to a Notion database. It includes scenarios to watch for new emails in Gmail, create corresponding entries in Notion, and update those entries based on actions taken in the GPT. This integration ensures that email data is efficiently transferred and organised for processing within the GPT.

Why is Notion used as an intermediary step for email management in this system?

Notion serves as a centralised database to store and organise emails. This approach is preferred over directly pulling from the inbox because it simplifies data retrieval for the GPT and allows for better tracking of email status (e.g., 'not reviewed', 'replied'). Using Notion enhances the system's ability to manage and query email data effectively.

Outline the two main Make.com scenarios created in the video and their respective purposes.

The first Make.com scenario watches the user's Gmail inbox for new, unread emails and automatically creates new entries in a designated Notion database, including details like subject, sender, and body. The second scenario is triggered by the GPT via a webhook, searches the Notion database for 'not reviewed' emails, and sends this information back to the GPT. These scenarios ensure seamless email data flow and management.

What is the function of the webhooks in facilitating communication between the GPT, Make.com, and Zapier?

Webhooks act as communication bridges between the different platforms. The Make.com webhook receives a request from the GPT to fetch emails from Notion. The Zapier webhook receives a 'post' request from the GPT containing the drafted reply details, which then triggers the email sending and Notion updating actions in Zapier. Webhooks enable real-time data exchange and automation across these platforms.

Explain the difference between the 'get' request used for fetching emails and the 'post' request used for sending replies.

A 'get' request is used to retrieve information from a server. In this case, the GPT sends a 'get' request (via the Make.com webhook) to fetch the 'not reviewed' emails from the Notion database. A 'post' request is used to send data to a server to create or update a resource. Here, the GPT sends a 'post' request (via the Zapier webhook) containing the reply details to Zapier, which then sends the email and updates Notion. Understanding these requests is crucial for setting up seamless data interactions.

What role does the schema play when setting up the actions within the ChatGPT GPT builder?

The schema defines the structure of the data that the ChatGPT action can expect to send and receive from external services like Make.com and Zapier. It specifies the API endpoint, request parameters, and response format, allowing ChatGPT to interact correctly with these platforms. Proper schema setup ensures accurate and reliable data exchanges during automation.

Describe the purpose of the Zapier automation in the email replying process.

The Zapier automation is responsible for taking the drafted email reply information sent by the GPT (via a webhook), creating a draft reply in the user's Gmail account in the correct email thread, and then updating the corresponding email's status in the Notion database to 'replied'. This automation streamlines the email reply process and keeps the Notion database updated.

What are the customisation possibilities of this automated GPT?

The automated GPT can be customised in various ways to better reflect the user's communication style and handle specific email types. User instructions, knowledge files, and tailored actions can be incorporated to enhance its functionality and personalisation. This flexibility allows users to adapt the system to their unique needs and preferences.

What are the potential challenges or obstacles when implementing this automated email system?

While setting up this system offers numerous benefits, users might encounter challenges such as technical complexity, integration issues, or data privacy concerns. Technical skills are required to configure webhooks, APIs, and automation platforms. Additionally, ensuring data security and compliance with privacy regulations is crucial when handling email data.

How does AI-powered email management impact productivity and communication workflows?

AI-powered email management can significantly enhance productivity by automating routine tasks and freeing up time for more strategic activities. It also improves communication workflows by providing consistent and well-crafted responses. However, users should consider ethical implications, such as data privacy and the potential for over-reliance on automation, which may impact personal touch in communication.

What are some practical applications of this automated email GPT in a business setting?

This automated email GPT can be used in various business scenarios, such as customer support, sales, and internal communications. For instance, customer support teams can use it to quickly draft and send responses to common inquiries, while sales teams can automate follow-up emails to prospects. Its versatility makes it a valuable tool across different departments.

What are some common misconceptions about using an automated GPT for email management?

Some common misconceptions include the belief that the GPT can fully replace human judgment or that it can operate without supervision. While the GPT can automate many tasks, users should still review and approve responses to ensure accuracy and appropriateness. Additionally, the system requires careful setup and monitoring to function effectively.

What are the strengths and weaknesses of integrating multiple platforms (ChatGPT, Make.com, Notion, Zapier) in this system?

Integrating multiple platforms allows for a comprehensive and flexible email management solution. The strengths include enhanced functionality, seamless data flow, and customisation options. However, potential weaknesses include increased complexity, dependency on multiple services, and the need for technical expertise to maintain the system.

What future developments or improvements could enhance this automated email system?

Future developments could include enhanced AI capabilities for more nuanced email drafting, improved integration with additional platforms, and more robust data security measures. Continuous updates and advancements in AI technology will likely drive further improvements in the system's efficiency and effectiveness, making it even more valuable for users.