Qualify inbound leads automatically with AI
Qualify inbound leads across many dimensions with AI and Zapier.
2024-11-13
If you’re receiving inbound sales leads, it’s important to qualify them so that you or your sales team aren’t taking low-value meetings. Previously, to do this, you might’ve had to manually sift through each lead to qualify them, outsource the task, or purchase expensive SaaS software to automate this process.
But now you can automate lead qualification across numerous criteria with AI, and you can set up this automated process in minutes with zero coding required.
In this tutorial, we’ll show you how you can automate this entire process using OpenAI and Zapier. We’ll take a lead submission from a web form, pass it through several AI steps, and output a qualification status at every step based on custom criteria, which you can then route to your CRM or other internal software for notifying your sales reps accordingly.
This tutorial was inspired by Siqi Chen’s demo from the Zapier AI Showcase.
You’ll need:
- Zapier paid plan
- OpenAI API
- Google Forms (or a form builder of your choice)
Steps:
- Step 1: Create a Google Form and connect it to Zapier
- Step 2: Generate an OpenAI key
- Step 3: Qualify the email domain with AI
- Step 4: Qualify the company with AI
- Step 5: Qualify the inbound request with AI
Step 1: Create a Google Form and connect it to Zapier
To start, create a Google Form to capture your inbound leads. On our form, we’re going to add fields for Name, Email, Company, and Message.
.webp)
Once your form is created, you can embed this on your website or link out to it from your landing page to start capturing leads.
Next, head over to Zapier to create a Zap automation that will ingest the leads from your form. To do this, go to Zapier, click the Zaps menu item on the left side, and click the Create button.
.webp)
For the trigger of your Zap, search and select your form builder. Since we’re using Google Forms, we’ll add Google Forms as the trigger.
We’ll set our Event to New Form Response so that our Zap triggers every time a new lead submits our Google Form.
.webp)
You’ll need to authenticate your Google Forms account in the following tab, then select your lead form, and finally send a test submission from your Google Form so it appears in the Test tab of your Zap step.
.webp)
Step 2: Generate an OpenAI key
Next, we’ll need to generate an OpenAI key. This will be used in the following step within our Zap to qualify our leads. To do this, log in to platform.openai.com, then click ‘API’ to be taken to the developer platform.
.webp)
Click ‘API keys’ on the left menu, and click ‘Create new secret key’.
.webp)
Give it a name and click ‘Create secret key’. Once generated, copy it to your clipboard. Also, keep it secure; do not share it publicly as it gives access to your OpenAI account capabilities.
Step 3: Qualify the email domain with AI
Now that we have our OpenAI key, we can connect it to our Zap and start using ChatGPT to qualify our inbound leads across a multitude of dimensions.
To get started with this, add a step to your Zap, search for ChatGPT, and set the Event to Conversation.
.webp)
This is where you’ll need your OpenAI API key. On the Account tab, when you click Connect, a pop-up window will appear where you can paste in your OpenAI API key. Drop it into the top text box and then click Continue.
.webp)
In the Action tab, we are going to set our first lead qualification prompt. For this first one, we want to disqualify leads that have a free email provider service (e.g. Gmail, Yahoo, etc.).
Sample prompt:
Email address: [Insert Email Address From Step 1 of the Zap]Examine the email address provided and determine whether it is from a well known free email provider like gmail.com, yahoo.com, qq.com, outlook.com and the like. Emails from free email provider domains are to be considered Unqualified. Emails from other domains should be marked as Qualified.
If you're not sure if the domain is a popular free email service, mark the lead as Qualified.
If you determine it is a qualified lead, you will return the string: "Qualified Lead". Otherwise, you will return the string: "Unqualified Lead".
.webp)
To finalize this qualification step, we’ll add a Filter by Zapier step to filter out any leads that are not qualified. To do this, add a Filter by Zapier step and set the logic to “2. Reply (Text) Contains Qualified Lead”. This is checking the output Reply from our ChatGPT step to make sure it contains “Qualified Lead,” if it does, the lead will continue to the next step, if it doesn’t, the Zap will complete at this step.
.webp)
Step 4: Qualify the company with AI
Now that we’ve qualified the email domain of our lead, we can qualify the company to make sure it meets our sales criteria. For this tutorial, our company qualification criteria will be companies that have at least $1M ARR, at least 100 employees, and are based in the United States.
We’ll also have this AI step generate a summary of the company for our sales team.
To do this, add another ChatGPT step with the Conversation Event.
Sample prompt:
Email: [Insert Email Address From Step 1 of the Zap]
Company Name: [Insert Company Name From Step 1 of the Zap]
You will examine the Email Domain and Company Name provided, and provide useful context to the sales team about this lead in a very short, very succinct paragraph of plain English of how qualified the lead is, and why.
To the extent that information is available, leads that are qualified will always have at least 100 employees and $1M in ARR, and be based in the United States. If they have less than 100 people or headcount information isn’t available, they should be disqualified immediately.
If you determine it is a qualified lead, you will begin your summary with the string "Qualified Lead: ". Otherwise, you will begin your summary with the string "Unqualified Lead: ". Do not return the quotes, only the string, then followed by your summary.
Test the output of this step to make sure it matches your expectations.
.webp)
If the output looks good, add another Filter by Zapier step and set it up exactly like the previous Filter by Zapier step, except that the input variable is the Reply from Step 4 of the Zap (the previous AI step).
.webp)
Step 5: Qualify the inbound request with AI
Now that we’ve qualified the email domain and the company, we can perform our final AI qualification step. For this, we’ll set criteria that we only want to qualify leads who are interested in our Enterprise product offering. We can do this by referencing the inbound lead’s message from our form for mentions of our Enterprise product offering.
To do this, add another ChatGPT step with the Conversation Event.
Sample prompt:
Inbound Request: [Insert Inbound Request from Step 1 of the Zap]
You will examine the Inbound Request provided, and provide useful context to the sales team about this request in a very short, very succinct paragraph of plain English of how qualified the lead is, and why.
To the extent that information is available, leads who are qualified will be interested in our Enterprise product offering. If they are not interested in our Enterprise product offering, then they are considered an Unqualified Lead.
If you determine it is a qualified lead, you will begin your summary with the string "Qualified Lead: ". Otherwise, you will begin your summary with the string "Unqualified Lead: ". Do not return the quotes, only the string, then followed by your summary.
.webp)
Test the output of this step to make sure it matches your expectations. If the output looks good, add another Filter by Zapier step and set it up exactly like the previous Filter by Zapier steps, except that the input variable is the Reply from Step 6 of the Zap (the previous AI step).
Turn on your Zap when you’re done.
.webp)
And that’s it! If you made it this far, you’ve set up a multi-step automated lead qualification process, checking for a proper email domain, specific company criteria, and inbound lead product interests. From here, you could add further steps to the Zap to route qualified leads to your CRM or internal communication channels, or even send an automated email, drafted by AI, to the lead to set up a time for a call.
This tutorial was created by Garrett.