In today’s fast-paced digital world, managing your inbox can feel overwhelming.

With constant incoming messages, it’s easy to lose track of important emails.

That’s where an AI-powered email triager can help. Using Lindy, you’ll create an automated system that intelligently sorts and labels your emails based on their content, allowing you to focus on what truly matters.

This tutorial will walk you through the steps of setting up an AI assistant to streamline your inbox management. By the end, you'll have a fully functioning email triager that reduces manual email sorting and keeps your inbox organized.

Key Objectives:

• Set up an AI email triage workflow using Lindy.
• Implement conditional logic for automatic email labeling.
• Create custom labels based on email content.
• Test and refine the email triage system.

Setting up the email triage workflow

First, create an account on Lindy.

Then you’ll need to go to Lindy’s marketplace and select the Email Triager option.

This pre-built template will provide the foundation for our AI assistant.

Configuring the Email Trigger:

The first element in the workflow is the Email Received trigger. Connect and authorize your Gmail or Outlook account to begin.

Choose whether to track follow-up emails or process only the first message in a thread.

Set up filters to specify which emails should trigger the workflow—this could be emails from specific contacts or departments. Honestly this really depends on what types of emails you want to filter. In my case I won't use any filters at all as I want to classify every incoming email.

Pro Tip: If you want to process only high-priority emails, consider filtering by keywords or sender addresses to ensure the right emails are triaged.

Creating conditional logic for labeling emails

Now that the basic flow is set up, it’s time to add logic that determines how emails will be categorized. Lindy uses AI prompts for easy setup:

Default Conditions:

The template includes basic conditions like high priority, spam, and newsletters. You can customize these or add new ones as needed. In my case, I'll use GPT-4o Mini as I think it's more than sufficient for this use case.

Adding Custom Conditions:

Let’s create a new condition for emails requesting meetings. Use a prompt such as “a customer is requesting a meeting” and name the condition “Meeting Request”.

Pro Tip: Lindy allows you to write these conditions using natural language, making it easier to create complex sorting rules.

Defining label names and actions

Next, we need to add the “Add Label to Email” step for our new condition. To do this, click the “Add Step” button coming from your new Condition.

Then select Perform an action

Then select the Gmail action Add Label to Email

After each condition, add a step to label the email. You can allow Lindy to automatically generate labels, or you can define them manually.

Pro Tip: You can use Lindy’s AI Prompt option to dynamically create labels based on the email’s content for more personalized sorting.

Testing and saving your workflow

With everything in place, it’s time to test your new triage system:

Save Your Workflow:

Testing:

Send a test email to your inbox and observe how Lindy processes and labels the email. For instance, if the email looks like a newsletter, it should be labeled “Newsletter”.

Conclusion

Congratulations! You’ve successfully built an AI-powered email triage assistant that automatically labels incoming emails based on their content. Over time, this will help you manage your inbox more effectively and reduce the manual effort of sorting through messages.

As your email needs evolve, remember to revisit and update your conditions and labels to maintain optimal efficiency.

‍