Automated Client Review Request (Zapier Agent) - Jason Pantana: AI + Marketing Training
Bot

Automated Client Review Request (Zapier Agent)

AiM Resources

What This Is

This is a prompt designed specifically for Zapier’s Agent Beta—a new feature that lets you describe your automation in plain language instead of building step-by-step Zaps. Rather than manually configuring each trigger, action, and filter, you simply write a prompt like the one below. The AI will then figure out which connected apps and functions to use to carry out your instructions.

In this example, the prompt triggers when a deal is marked “Closed” in a Google Sheet. However, you can modify the trigger to use any app supported by Zapier—such as Monday, Dotloop, or Follow Up Boss. Just adjust the wording in your prompt accordingly.

The workflow then scans client emails for sentiment. If there’s no negative tone, it sends a personalized review request with links to key platforms.

Note: This is a beta feature, so it may have bugs or limitations. You may need to test and refine your prompt to ensure consistency.

This represents a broader move toward agentic AI—where AI tools take action on your behalf using natural language instructions, automating tasks in a more flexible, intelligent way than traditional rule-based systems.

How to Use It

  1. Enable the Agent Beta in your Zapier account (access may be limited—check your account or request it if needed)
  2. Create a new Agent by tapping the “+ New Agent” button in your Zapier dashboard
  3. Paste your prompt (like the one provided below) into the Agent’s instruction field
  4. Connect the required apps when prompted (e.g., Google Sheets, Gmail)
  5. Click the gear icon next to a connected app to manually map specific data or settings—otherwise, the AI will make its best guess
  6. Test the Agent to confirm it understands and executes the correct actions
  7. Review the output and adjust your prompt as needed for tone, accuracy, or logic (the Agent learns from your chat)
  8. Activate the Agent and monitor how it performs in real scenarios

Note: You’ll want to modify the bold text in the prompt below to fit your own apps, links, or phrasing. Treat it as a customizable template—not a one-size-fits-all solution.

When the “Deal Status” column of a row in a Google Sheet is changed to “Closed”.
1. Scan all email folders (inbox, sent, spam, trash, and subfolders) associated with the client(s) for sentiment related to the agent and the transaction Gmail: [Gmail: Find Email]
If negative sentiment is detected in any email:
– Stop the workflow and output, citing: “Negative sentiment detected – email not sent.”
– If sentiment is neutral, positive, or not applicable:
2. Send a first-name personalized, well-formatted email to the client(s): [Gmail: Send Email]
– Address the client(s) by first name.
– Reference the property address and one highlight from the deal details found in the Google Sheet. If no meaningful detail is available, don’t make something up—just skip this part.
– Clearly indicate whether the client purchased or sold a home based on the Google Sheet data.
– Include the following review links:
> Google: [INSERT YOUR LINK]
> Zillow: [INSERT YOUR LINK]
> Yelp: [INSERT YOUR LINK]
> Facebook: [INSERT YOUR LINK]

– Ensure the email tone is warm, grateful, and concise (around 150–200 words).
– Send the email directly to the client(s) using their contact information
End the email with the following sender signature:
Best regards,
[INSERT YOUR NAME]


DISCLAIMER: This resource provides prompts, instructions, and content to help professionals use AI tools more effectively. Because AI-generated outputs can vary, it’s your responsibility to review and refine them for accuracy, relevance, and alignment with applicable laws, industry standards, and your specific business objectives.