Back to list

Building Custom GPT Powered by HubSpot in Slack

Learn how to build a custom GPT chatbot in Slack, integrated with HubSpot’s CRM to streamline workflows and access customer insights. This guide covers setup, integration, and practical use cases for improving team productivity.

In sales, customer support, and business development, quickly reviewing past conversations is crucial for delivering personalized service. However, finding these details often requires multiple steps—opening a browser, navigating HubSpot, and searching through interactions—wasting time and diverting focus from client engagement.

What if you could streamline this process? Imagine using familiar tools like Slack or Teams to instantly summarize past conversations, enabling faster, more effective customer support.

The Benefits of Integrating Your Tools with AI

By leveraging AI-driven tools to summarize your recent conversations, you can experience several key benefits:

  • Increased Efficiency: By staying within your current toolset, like Slack, you can catch up on essential details without constantly switching platforms.
  • Simplified AI Adoption: Both you and your colleagues can benefit from AI without getting lost in the technical complexity or needing specialized training.
  • Optimized Workflows: You can configure the AI to suit your unique workflows, enhancing overall productivity.

How to Integrate Your Tools

There are multiple ways to integrate chat tools with platforms like HubSpot utilizing AI capabilities:

  1. Using Services like Runbear and Make (in 30 minutes): These platforms offer easy-to-use interfaces to set up integrations without needing extensive programming knowledge.
  2. Developing the System from Scratch (2+ months): For those with technical expertise, building a custom solution might be the way to go, allowing maximum flexibility and customization.
  3. Hiring AI Consultants ($100k+): If resources allow, hiring experts can accelerate the integration process and ensure that the system meets your business requirements comprehensively.

Step-by-Step Guide: Use Runbear and Make in 20 Minutes

Here's a quick guide on how you can set up a system in just 30 minutes using Runbear and Make:

Step 1: Build Scenarios on Make.com

Start by creating scenarios on Make.com that can fetch customer data from HubSpot. This setup will automate the data collection process, ensuring that you always have the most up-to-date information available.

  1. Use this HubSpot Conversations API Make Template to create a simple scenario. The scenario accepts webhook requests, searches for contacts, and responds to email conversations.
  2. Click the first step and click the “Copy address to clipboard” to copy the URL. Keep the URL in somewhere.

Check out this interactive guide for details:

Disclaimer: This template does not authenticate the requests. Fortunately, you can easily add authentication steps using Make and Runbear. Contact Runbear team to learn more!

Step 2: Create a Custom GPT in Slack Using Runbear

With Runbear, you can set up a custom GPT-powered assistant within Slack. This will allow you to generate summaries of past conversations and data insights without leaving the chat window.

Create a new assistants and add Use function calling to fetch the necessary data from HubSpot. at the end of your instructions. For example,

You are a professional sales/marketing/customer support assistant. Your role is to provide precise and helpful responses to customer inquiries, assist in lead qualification, and support the sales process. Use a professional tone in all communications. If data or information is unavailable, admit that you don't know. Always aim to provide clear, concise responses. Use function calling to fetch the necessary data from HubSpot. 

Click the “Add OpenAPI” button and enter this OpenAPI JSON spec. Don't forget to replace the {your_make_host} and {your_make_path} with your value.

For example, if your webhook URL is https://hook.us1.make.com/abcdefg1234567890hijklmnopqrstuv, replace {your_make_host} with https://hook.us1.make.com and {your_make_path} with abcdefg1234567890hijklmnopqrstuv. Be sure to keep the slashes (/) before and after {your_make_path} intact.

{
  "openapi": "3.1.0",
  "servers": [
    {
      "url": "{your_make_host}",
      "description": "Make.com server"
    }
  ],
  "paths": {
    "/{your_make_path}/": {
      "get": {
        "summary": "Get conversations with the contact",
        "description": "Get conversations with the contact",
        "operationId": "get_hubspot_conversations",
        "parameters": [
          {
            "name": "query",
            "in": "query",
            "schema": {
              "type": "string",
              "description": "Contact information (e.g., email, name, company)"
            },
            "required": true
          }
        ]
      }
    }
  }
}

Create the assistant and connect it to you Slack workspace.

Step 3: Enjoy Your Custom HubSpot Assistant in Slack

With everything set up, you can now enjoy the seamless experience of accessing customer information and conversation summaries right from Slack. This integration can help you stay informed and reactive, enhancing client interactions without the need for constant context-switching.

Embrace the power of AI

Don't fall behind in the AI revolution. Start integrating these advancements into your workflows today and position yourself as an industry leader. Embrace the power of AI, and transform the way you and your team engage with clients and customers.

Begin your journey towards optimized customer interactions and lead your industry by setting up your AI-powered integration today. Your future self will thank you.