AI Agent for Dream 100 Prospecting using Perplexity and Google Sheets - n8n Workflow

• Sales and marketing professionals needing highly targeted lead generation.


• Business owners looking to define their ideal customer profile (ICP) and top prospects.


• Users interested in advanced LangChain-based n8n templates and AI agent orchestration.


• Anyone seeking a powerful, automated n8n workflow for market research.

Manually researching 100 ideal prospects—your 'Dream 100'—is incredibly time-consuming. This specialized n8n workflow solves this by creating a research AI Agent that automates the entire process. This sophisticated n8n template uses a conversational approach, allowing users to define their target criteria interactively. The core of this n8n automation leverages Perplexity AI's powerful research capabilities to find high-quality prospects that match a detailed profile. The agent then systematically writes the findings, including prospect details, summaries, and reasons for fit, directly into a Google Sheet. This saves countless hours of manual work and provides a consistently structured list, making it one of the most effective n8n templates for B2B prospecting.

This comprehensive n8n workflow begins with an input, typically a When chat message received n8n trigger or a Slack message.

1. Input Handling: The initial user prompt defining the ideal customer profile (ICP) is captured. A Switch n8n node determines if the input is a fresh user request or an internal re-trigger signal used for pagination.


2. AI Agent Execution: The request is passed to the core AI Agent n8n node, which uses an OpenRouter Chat Model (the LLM) and Simple Memory to maintain context. The agent is strictly instructed to collect all necessary ICP details before starting research.


3. Research and Storage Tools: The agent uses two primary n8n tools: Research via Perplexity to find prospects matching the criteria and the Insert Data Into Sheet n8n node (Google Sheets Tool) to structure and append the findings.


4. Pagination Logic: After the agent deposits a batch of names (e.g., 10), the n8n workflow checks the current status of the Google Sheet using the Check First Row and Check Last Row n8n nodes (looking for Prospect ID 1 and Prospect ID 100).


5. Continuation or Completion:

To deploy this powerful n8n workflow template, follow these steps:

1. Import: Copy the provided JSON data and paste it into your n8n canvas using the 'New' -> 'Import from JSON' option.


2. Google Sheet Setup: Create a copy of the provided Google Sheet template (URL is often noted in the sticky notes within the n8n template). Ensure your sheet has the necessary columns: 'Prospect ID', 'Prospect Name', 'Prospect Summary', and 'Reason'. Update the credentials and document ID in all Google Sheets and Insert Data Into Sheet n8n nodes.


3. Credential Configuration:

1. Activation: If using the Slack trigger, configure the When Slack Received n8n trigger and update the output Slack n8n nodes (Collect More Info Slack, All Done Slack) for proper messaging. Activate the main n8n workflow to begin.

When chat message received (n8n trigger): Initiates the n8n workflow upon receiving a message, typically from the internal n8n chat UI, which is ideal for testing.
AI Agent (LangChain Agent n8n node): The core intelligence. It coordinates research and data entry based on an extensive system prompt detailing the 'Dream 100' criteria and the required output format for the Google Sheet.
OpenRouter Chat Model (n8n node): Provides the language model capabilities for the agent to understand context, plan actions, and structure output.
Research via Perplexity (n8n node): Serves as the web research tool, enabling the agent to perform deep searches and find relevant prospects.
Insert Data Into Sheet (Google Sheets Tool n8n node): Used by the AI agent to append structured data (ID, Name, Summary, Reason) directly into the target Google Sheet.
Check First Row / Check Last Row (Google Sheets n8n node): Reads specific rows (Prospect ID 1 and 100) to determine if the target list is empty, partially filled, or complete.
Agent Re-trigger chat message (Set n8n node): Holds the specific, automated prompt used to instruct the AI Agent to continue researching the next sequential batch of prospects (e.g., prospects 21-30), preventing duplicates and driving continuous research. This is vital for the looping structure of the n8n workflow.
Do we need MORE data in the sheet? (If n8n node): Implements the pagination logic, deciding whether to re-inject the 'Re-trigger Chat' message back into the Agent loop or mark the n8n workflow as finished.