LinkedIn Lead Qualification n8n Workflow: Apify + Unipile + OpenAI

Discover, qualify, and store LinkedIn leads automatically—daily—using Apify, Unipile, and OpenAI inside n8n

This LinkedIn Lead Qualification n8n Workflow: Apify + Unipile + OpenAI runs on a schedule to find LinkedIn profiles, enrich them with Unipile (including recent posts), then uses an OpenAI model to produce a JSON qualification decision and score. Leads marked as accepted are inserted into your Nebula Postgres table for downstream CRM/engagement.

What this workflow does

• Schedules a daily run via a schedule trigger at your chosen hour.
• Scrapes LinkedIn profiles with Apify by starting an Apify LinkedIn profile search actor, then loads the discovered profiles dataset.
• Filters low-signal leads (e.g., students/job seekers) and keeps only profiles with sufficient audience size and relevant AI/automation keywords.
• Enriches each lead one-by-one by extracting the LinkedIn username, fetching full profile details and the latest posts from Unipile.
• Qualifies with OpenAI by combining the profile context with the last four posts and sending it to OpenAI to return: decision, score, categories, and a summary in JSON.
• Saves accepted leads into Nebula Postgres and flags them to store in your CRM, then continues looping through remaining profiles.

Use cases

• SaaS operators who want an automated inbound lead qualification pipeline from LinkedIn.
• Automation engineers building a CRM-ready lead database without manual research.
• Teams targeting AI/automation decision-makers and wanting post-aware qualification signals.

Technical details

• Nodes / integrations: schedule trigger, Apify (via HTTP request), if, code, wait, postgres, plus supporting notes.
• Requires Apify API token, Unipile API credentials (including account_id), and an OpenAI (or compatible) API credential.
• Inserts accepted profiles into your Nebula Postgres schema/table (update the target table/fields).