Automated Local Lead Generation with n8n: Building an AI ...

Quoted on this wiki

Every place a page here uses this source, in the order the words come in it.

1. Next, we want the agent to return a structured item of found leads, to insert them into Google Sheets. For this, I have defined the following schema, which is found inside the "Parse Output" node: “For traceability, I have included the "google_places_query" that the agent used to query Google Places, and a language_code, as this improves the enrichment quality (querying in the language of the lead) of Part 2.” After this, I have a code node that does a small conversion from the output node, to directly return the array within the output property:

   In Enrichment APIs and integrations

2. Trigger when a new lead is added to the lead list on Google Sheets “It will scrape different URLs and only stop when it has gathered enough information on the lead.” Below, I will elaborate on the most important nodes for Part 1 of the workflow.

   In Waterfall enrichment

3. Trigger when a new lead is added to the lead list on Google Sheets “This is where an LLM really shines; it allows for classification of data based on unstructured text.” Below, I will elaborate on the most important nodes for Part 1 of the workflow.

   In Lead enrichment workflows

4. Trigger when a new lead is added to the lead list on Google Sheets “It will scrape different URLs and only stop when it has gathered enough information on the lead.” Below, I will elaborate on the most important nodes for Part 1 of the workflow.

   In Lead enrichment workflows

5. -- “Although agentic frameworks like LangGraph are really powerful, they come with added complexity making it harder to develop and maintain them.” Also, what makes this workflow unique is its tool use; Many tutorials, my previous LangGraph guide included, use top-level wrappers like Serper to fetch leads, which are usually quite expensive if you want to gather a fair amount of leads. In this guide however, I directly use the Google Places API, allowing for practically free scraping of thousands of leads (GCP has a really generous free tier). While this adds some setup complexity, the trade-off is well worth it given the cost and flexibility benefits, allowing for powerful local lead scraping.

   In Lead enrichment workflows

Quotes are verbatim and belong to their publishers. Where a quote has been translated, the original wording is kept below it.